Re

fe
Pl
r en
ea c
se eM
Do

Network+ Certification
No at
tC er
op
y ial
Network+ Certification

ial
To access your Online ANYTIME Learning Components, please go to
http://onlineanytime.newhorizons.com and use the username and password
provided to you in class.

er
WBT Username:

WBT Password:

at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re
 NETWORK+ CERTIFICATION
Course Number: NH77700 (0751EBEE)
Course Edition: 3.2
For software version: na

ial
ACKNOWLEDGEMENTS
Project Team
Curriculum Developers and Technical Writers : Gail Sandler and Pamela J. Taylor • Copy Editor : Taryn
Chase • Reviewing Editor : Tom Elston • Layout Technicians : Kristi Toscano and Isolina Salgado • Quality

er
Assurance Analyst : Frank Wosnick

Project Support
Managing Editor, Internet & Programming : Carlene Jo Kline • Managing Editor, Personal Computing and
Design : Cheryl Russo • Managing Editor, IT Professional : Susan B. SanFilippo • Manager of Instructor-

at
Led Web Curriculum : Joy Morris • Managers of Web Development : Joseph Blakely and Tracy Dedes •
Manager of Publishing Services : Mark Onisk • Publishing Services Supervisor : Rachel Miller • Publishing
Technical Specialist : Daniel L. Quackenbush • Instructional Design Specialist : Susan L. Reber • Project
Technical Support Specialist : Edward Hyder • Testing & Layout Technician : Bonnie S. Watts

y
Administration

op
eM
Vice President of Content & Operations : Paul A. Krause • Director of Advanced Technology Content : Kent
Michels • Director of Personal Computing and Design Content : Joy Insinna • Director of Interactive
Content : Jeff Felice • Director of Online Education : Lesley Darling • Director of Operations & Publishing
Services : Kress Riley

NOTICES tC
No
DISCLAIMER: While Element K Content LLC takes care to ensure the accuracy and quality of these materials, we cannot guarantee their accuracy, and all materials are
c
provided without any warranty whatsoever, including, but not limited to, the implied warranties of merchantability or fitness for a particular purpose

TRADEMARK NOTICES: Element K Content LLC, ZDU, FirstEdition, Element K Journals and the corresponding logos are trademarks and service marks of Element K
Content LLC. Windows NT and NetWare are registered trademarks of Microsoft, Inc., and Novell, Inc., respecitvely. All other product names and services used throughout this
book are common law or registered trademarks and service marks of their respective companies. Use of another entity’s product name or service in this book is for editorial
en

purposes only. No such use, or the use of any trade name, is intended to convey endorsement or other affiliation with the book.
Do

Copyright © 2000 Element K Content LLC. All rights reserved. This publication, or any part thereof, may not be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, storage in an information retrieval system, or otherwise, without express written permission of Element K
Content LLC, 500 Canal View Boulevard, Rochester, NY 14623, (800) 434-3466. Element K Content LLC’s World Wide Web site is located at www.elementkcourseware.com

Unauthorized reproduction or transmission of any part of this book or materials is a violation of federal law. If you believe that this book, related materials, or any other
Element K Content LLC materials are being reproduced or transmitted without permission, please call 1-800-478-7788.
se
r
fe
ea
Pl
Re

ii
NETWORK+ CERTIFICATION CONTENT
OVERVIEW

ial
About This Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Lesson 1: Basic Networking Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

er
Lesson 2: Servers and Clients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Lesson 3: Introduction to the OSI Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

at
Lesson 4: The Physical Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

y
Lesson 5: The Data Link Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

op
eM
Lesson 6: Data Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Lesson 7: The Network Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

tC
Lesson 8: The Transport Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
No
Lesson 9: The Session Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
c

Lesson 10: TCP/IP Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

en
Do

Lesson 11: IP Addressing, Subnetting, and Supernetting . . . . . . . . . . . . . . . . . . . . 259

Lesson 12: Accessing and Using a TCP/IP Network . . . . . . . . . . . . . . . . . . . . . . . 289
se
r

Lesson 13: Remote Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

fe

Lesson 14: Network Security and Fault Tolerance . . . . . . . . . . . . . . . . . . . . . . . . 337

ea

Lesson 15: Preparing for Network Installation . . . . . . . . . . . . . . . . . . . . . . . . . . 381

Pl
Re

Lesson 16: Maintaining the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

Lesson 17: Troubleshooting the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423

Contents iii
CONTENT Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

OVERVIEW Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525

ial
er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

iv
NETWORK+ CERTIFICATION CONTENTS

ial
About This Course
About This Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Course Setup Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

er
How to Use This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

LESSON 1: BASIC NETWORKING KNOWLEDGE

at
TOPIC 1A Network and Internetwork Concepts . . . . . . . . . . . . . . . . . . 2

y
TOPIC 1B Network Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

op
TOPIC 1C
eM
Networking Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Lesson 1 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

LESSON 2: SERVERS AND CLIENTS

TOPIC 2A Network Operating System Basics . . . . . . . . . . . . . . . . . . . . 18
tC
No
c

TOPIC 2B Network Clients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Apply Your Knowledge 2-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
en

TOPIC 2C Directory Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Do

Check Your Skills 2-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Lesson 2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

LESSON 3: INTRODUCTION TO THE OSI MODEL

se
r
fe

TOPIC 3A Introduction to the OSI Model . . . . . . . . . . . . . . . . . . . . . . . 50

ea

Apply Your Knowledge 3-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

TOPIC 3B Families of Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Pl
Re

Lesson 3 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Contents v
CONTENTS LESSON 4: THE PHYSICAL MEDIA
TOPIC 4A Data-transmission Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Apply Your Knowledge 4-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

TOPIC 4B

ial
Unbounded Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

TOPIC 4C Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

TOPIC 4D Signal Transmissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Lesson 4 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

er
LESSON 5: THE DATA LINK LAYER
TOPIC 5A

at
Data Link Layer Considerations . . . . . . . . . . . . . . . . . . . . . .100

TOPIC 5B Media Access Control (MAC) and Logical Link Control

y
(LLC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101
Check Your Skills 5-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

op
eM
TOPIC 5C IEEE 802 Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110
Check Your Skills 5-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

tC
Lesson 5 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

LESSON 6: DATA TRANSMISSION

No
c

TOPIC 6A Introduction to Network Connection Devices . . . . . . . . . . .130

Check Your Skills 6-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Apply Your Knowledge 6-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
en
Do

TOPIC 6B Network Adapters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .142

Apply Your Knowledge 6-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Apply Your Knowledge 6-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

TOPIC 6C
se

Putting Data on the Cable . . . . . . . . . . . . . . . . . . . . . . . . . . .154

r

TOPIC 6D Network Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159

fe

Apply Your Knowledge 6-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

ea

TOPIC 6E Bridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169

Check Your Skills 6-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Pl
Re

Apply Your Knowledge 6-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Lesson 6 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

vi
LESSON 7: THE NETWORK LAYER CONTENTS
TOPIC 7A Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .188
Check Your Skills 7-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

TOPIC 7B

ial
Internetwork Connection Devices . . . . . . . . . . . . . . . . . . . . .190

TOPIC 7C Routers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .193

Apply Your Knowledge 7-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
Lesson 7 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

er
LESSON 8: THE TRANSPORT LAYER
TOPIC 8A Transport Layer Addressing Issues . . . . . . . . . . . . . . . . . . . .204

at
TOPIC 8B The Transport Layer Segmenting Process . . . . . . . . . . . . . . .207

TOPIC 8C

y
Transport Layer Connection Services . . . . . . . . . . . . . . . . . .209
Apply Your Knowledge 8-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

op
eM
Lesson 8 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

LESSON 9: THE SESSION LAYER

TOPIC 9A tC
Dialog Control Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216
No
TOPIC 9B Managing the Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .220
c

Lesson 9 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

en

LESSON 10: TCP/IP FUNDAMENTALS

Do

TOPIC 10A What is TCP/IP? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .224

TOPIC 10B The TCP/IP Protocol Suite . . . . . . . . . . . . . . . . . . . . . . . . . . .229

se
r

Apply Your Knowledge 10-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

Lesson 10 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
fe
ea

LESSON 11: IP ADDRESSING, SUBNETTING, AND SUPERNETTING

Pl
Re

TOPIC 11A IP Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .260

TOPIC 11B Interior Gateway Protocols . . . . . . . . . . . . . . . . . . . . . . . . . .268

TOPIC 11C Introduction to Subnetting . . . . . . . . . . . . . . . . . . . . . . . . . .279

Apply Your Knowledge 11-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284

Contents vii
CONTENTS TOPIC 11D Supernetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Lesson 11 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287

LESSON 12: ACCESSING AND USING A TCP/IP NETWORK

ial
TOPIC 12A Configuring Clients for TCP/IP Networks . . . . . . . . . . . . . . .290
Check Your Skills 12-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

TOPIC 12B TCP/IP Suite Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .294

er
TOPIC 12C Using TCP/IP Suite Utilities to Check Network Connectiv-
ity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .307
Apply Your Knowledge 12-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Check Your Skills 12-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
Lesson 12 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

at
LESSON 13: REMOTE ACCESS

y
op
TOPIC 13A
eM
Remote Access Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . .318

TOPIC 13B Examining Remote Connection Options . . . . . . . . . . . . . . . .321

TOPIC 13C
tC
Dial-up Networking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .326
Apply Your Knowledge 13-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Check Your Skills 13-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
No
Lesson 13 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
c

LESSON 14: NETWORK SECURITY AND FAULT TOLERANCE

en
Do

TOPIC 14A Security Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .339

TOPIC 14B Password Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347

TOPIC 14C
se

Data Encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349

r

TOPIC 14D Firewalls and Proxy Servers . . . . . . . . . . . . . . . . . . . . . . . . . .352

fe

Apply Your Knowledge 14-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

ea

TOPIC 14E System Fault Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . .355

Apply Your Knowledge 14-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
Pl
Re

Apply Your Knowledge 14-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371

TOPIC 14F Backing Up Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372

Apply Your Knowledge 14-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Lesson 14 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380

viii
LESSON 15: PREPARING FOR NETWORK INSTALLATION CONTENTS
TOPIC 15A Planning and Gathering Information . . . . . . . . . . . . . . . . . .382

TOPIC 15B Environmental Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .385

ial
TOPIC 15C Networking Components and Connectors . . . . . . . . . . . . . .389
Apply Your Knowledge 15-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394

TOPIC 15D Compatibility Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .396

Apply Your Knowledge 15-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398

er
Lesson 15 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

LESSON 16: MAINTAINING THE NETWORK

at
TOPIC 16A Network Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . .402

TOPIC 16B

y
Backup Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405

op
TOPIC 16C
eM
Virus Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .412
Apply Your Knowledge 16-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416

TOPIC 16D Patches and Other Software Updates . . . . . . . . . . . . . . . . . .417

tC
Apply Your Knowledge 16-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
Lesson 16 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
No
LESSON 17: TROUBLESHOOTING THE NETWORK
c

TOPIC 17A Exploring Troubleshooting Models . . . . . . . . . . . . . . . . . . . .424

en
Do

Apply Your Knowledge 17-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434

TOPIC 17B Identifying Troubleshooting Tools . . . . . . . . . . . . . . . . . . . .435

Apply Your Knowledge 17-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
Apply Your Knowledge 17-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
se
r

Lesson 17 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442

Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .443
fe
ea

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .513

Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .525
Pl
Re

Contents ix
Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
 ABOUT THIS
COURSE

ial
ABOUT THIS COURSE
Network+ Certification is an instruction book that will introduce
you to networking theory in a non-vendor-specific course.

er
Course Prerequisites
We designed Network+ Certification for the student with 18 to
24 months of professional computer support experience. It is assumed that you

at
have a general working knowledge of personal computers, some network experi-
ence, and have taken the following courses (or have equivalent experience):
Windows 95 - Level 1, or Windows 98 - Level 1.

y
op
Course Objectives
eM
When you’re done working your way through this book, you’ll be able to:

• Describe the purpose of networking protocols and networking in general.

•

•
to access them.
Describe the OSI networking model and its relationship to networking
tC
Identify features of various network operating systems and the clients used
No
components.
c

• Describe the network components relating to the Physical layer of the OSI
model.
• Describe the function of the Data Link layer of the OSI model.
en

•
Do

Explain how data is transmitted over a network.

• Describe the function of the Network layer of the OSI model.
• Describe the function of the Transport layer of the OSI model.
• Describe the function of the Session layer of the OSI model.
se

•
r

Explain the fundamental concepts of the TCP/IP protocol suite.

• Explain the use of TCP/IP addresses and subnets.
fe

• Access and use a TCP/IP network.

ea

• Describe the requirements for remote network access.

Pl
Re

About This Course xi

 •
ABOUT THIS Explain the need for and ways to implement network security and fault
tolerance.
• Describe the types of information that need to be gathered prior to installing
COURSE •
or updating a network operating system.
Explain ways to monitor and manage a network.
•

ial
Describe a systematic approach for troubleshooting network problems.

COURSE SETUP INFORMATION

er
Hardware and Software Requirements
While this course can be taught using only Windows 95/98 workstations for each
student and a classroom Windows NT 4 Server, the course is written to also

at
include a NetWare 4.11 server and optionally, a Linux server. Red Hat Linux 6.0
was used to develop this course.
To run this course, you will need:

y
• For each server:

op
—
eM
At least 32 megabytes (MB) of Random Access Memory (RAM)
— 1.2 GB hard drives
— A Pentium or faster processor
—
—
—
A mouse or compatible tracking device
A 3.5″ 1.44 MB floppy-disk drive
tC
A VGA or higher resolution video card and monitor
No
— A CD-ROM drive
c

— Windows NT Server 4.0 (with Service Pack 4 or higher applied)

— NetWare 4.11 (optional)
en

— Linux (optional)
Do

• For each workstation:

— At least 16 megabytes (MB) of Random Access Memory (RAM)
— 500 MB hard drives or larger
se

— A Pentium or faster processor

r

— A VGA or higher resolution video card and monitor

fe

— A mouse or compatible tracking device

ea

— A 3.5″ 1.44 MB floppy-disk drive

— A CD-ROM drive
— Windows 95
Pl
Re

— Internet access
— An email account for each workstation

xii
— Headphones or speakers (If students will be doing the video presentations
included on the Interactive Learning CD each from their own workstation,
issue each workstation a set of headphones. If you’ll be playing the video
ABOUT THIS
presentations and displaying them to students from the Instructor’s worksta-
tion, you’ll need speakers so students can hear the audio portion of the COURSE
presentations.) A sound will also be required to use the video presentations.

ial
— All systems require network cards and all systems are networked together.
(We used 10-Base T networking while developing this course.)

Class Requirements
In order for the class to run properly, perform the procedures described below.

er
Install and configure classroom Windows NT 4 server:
1. Partition and format the hard disk into a primary, DOS-bootable FAT
partition.

at
2. Install DOS 6.2x or higher on the FAT partition in the C:\DOS directory.

3. Create an AUTOEXEC.BAT file that includes at a minimum, the line

y
@path=C:\DOS.

op
4.
eM
Install Windows NT Server 4.0 as the PDC of NETDOMAIN. Name the
computer NTSRVR. Install NetBEUI and NWLink. Install TCP/IP and con-
figure it as required for your classroom. Don’t install Internet Information

tC
Server (IIS). Change the Administrator password to password. Create an
Emergency Repair Disk.

5. Install a printer named class. Share it as class. (Don’t install a print device.)
Pause the printer. Install a printer named class . Share it as class . (Don’t
No
install a print device.)
c

6. Create a user account called Domain_Admin with the password of manage,

uncheck User Must Change Password At Next Logon, and add the account
en

to the Domain Admins group.

Do

7. Create a user account for each student in the class (Stu1 through Stu#).

Install and configure the classroom NetWare 4.11 server:

se

1. Create a new server using the Simple Installation method.

r

2. Name the server 411SRVR.

fe
ea

3. The server will contain only one volume, SYS:.

4. The Context for this server and its objects should be CLASS.
Pl
Re

5. Accept defaults as appropriate.

Install and configure the classroom Linux server:

1. Boot from the Linux CD-ROM, and install using the Server Installation
Class.

About This Course xiii

ABOUT THIS 2.

3.
Accept defaults as appropriate.

Name the server linuxsrvr.

COURSE 4. Assign the password Passw0rd (with a capital P and the number 0) to the
root account.

ial
5. Configure TCP/IP as appropriate to your classroom.

6. Create a user account for each student in the class (user1 through user#).
You can use the adduser command to add the users (log in as root, then
enter adduser user#). After creating the users, use the passwd command to
assign each student a password.

er
Install and configure student and instructor workstations:
1. Format the computer hard drive so that it is running only DOS.

2. Run Windows setup. Select Typical installation. Allow Windows 95 to detect

at
your computer’s hardware. Select Network Adapter, if available. We
assumed the following settings when developing the course:
•

y
The floppy-disk drive is A.
• A Network Neighborhood is present. If you don’t have a Network

op
eM
Neighborhood, you may not be able to perform some of the tasks.

3. Select the option to install the most common Windows 95 components.

tC
4. Set the appropriate time zone.

5. Install the printer driver of your choice (this course was written using the HP
LaserJet 4Si printer driver). Name the printer Class Printer.
No
You don’t need a printer to
c

6. Configure TCP/IP as necessary for your classroom.

run the course, you just need
a printer driver. 7. Install Dial-up Networking. (In the Add/Remove Programs Control Panel
utility, display the Windows Setup page. Select Communications, and then
en

click Details. Select Dial-up Networking, and click OK. Accept defaults as
Do

This configuration eliminates

prompted.)
variances in user settings,
Start-menu groupings, DOS
environmental settings, and 8. Set up the Network+ Interactive Learning CD on each workstation by run-
device-driver conflicts that ning SETUP from the CD-ROM.
might result from upgrading
se
r

a previous version of 9. Install and configure both Client for Microsoft Networks and Novell Client
Windows. software on each workstation. For the Novell Client software, specify the
fe

default server as 411SRVR.

ea

10. Copy the CAB files to C:\CAB on each workstation, or make the Windows
95 CDs available for use if needed during class.
Pl

11. Log in to the NetWare server as Admin. In NetWare Administrator, create an

Re

Organizational Unit container under Class named NetPlus. Create some users
and a group under the NetPlus container.
If you have examples of any of the following which students can see (and possi-
bly examine in more detail), it would enhance students’ learning experience.
Items of special interest might include: NICs, patch panels, cables, repeaters,
hubs, bridges, routers, brouters, print services devices, UPS, and so forth.
xiv
If you have examples of any network media that students could view and possi-
bly manipulate, it would be beneficial to students. This might include: various
connector types such as RJ-45, RJ-11, BNC, IBM genderless, and IBM external
ABOUT THIS
transceivers; vampire taps; cable-making tools; and pieces of cable to create
coaxial and 10BaseT cables. COURSE
One of the optional activities in the course is to ″create a network without

ial
computers.″ This involves students locating a network card and the appropriate
cabling to go with it, any additional devices such as hubs or MSAUs, and attach-
ing them all together. If you have several different topologies that students could
choose from, it would make them think about what pieces really go together.

er
HOW TO USE THIS BOOK
You can use this book as a learning guide, a review tool, and a reference.

at
As a Learning Guide

y
Each lesson covers one broad topic or set of related topics. Lessons are arranged
in order of the OSI model, followed by topics that build on this knowledge. For

op
eM
this reason, you should work through the lessons in sequence.
We organized each lesson into explanatory topics and step-by-step activities. Top-
ics provide the theory you need to prepare for Network+ certification; activities
allow you to apply this theory to practical hands-on examples or for you to dis-

tC
cuss the concepts presented.

As a Review Tool
No
Any method of instruction is only as effective as the time and effort you are will-
c

ing to invest in it. For this reason, we encourage you to spend some time
reviewing the book’s more challenging topics and activities.
en
Do

As a Reference
You can use the Concepts sections in this book as a first source for definitions of
terms, background information on given topics, and summaries of procedures.
se
r
fe
ea
Pl
Re

About This Course xv

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Basic Networking LESSON
Knowledge
1

ial
Data Files:
none
Overview
Lesson Time:

er
This lesson introduces the concept of a network and how the network com- 45 minutes
puters are connected.

at
Objectives
To understand the purpose of the networking protocols and networking in general,

y
you will:

op
1A Define network concepts.
eM
This topic identifies and explains the roles played by basic network
components. It also defines interconnectivity and interoperability in terms

tC
of networking.

1B Differentiate between physical and logical topologies.

Now that you know what a network is, you’ll examine the physical topol-
No
ogy or structure of the network and the logical topology or path of a
c

signal on the network.

1C Define networking strategies.

en

You’ll examine two major types of networks—client-server and peer-to-

Do

peer.
se
r
fe
ea
Pl
Re

Lesson 1: Basic Networking Knowledge 1

 TOPIC 1A
Network and Internetwork Concepts
Network computing results from three basic needs: the need to share information;

ial
the need for a way, or path, of communicating with others; and the need for rules
to control how communication occurs. This course addresses these three basic
network computing needs.
A computer network is a collection of hardware and software that enables a group
of nodes (devices) to communicate and provides users with access to shared
resources. Nodes can be computers, such as mainframe computers, minicomput-

er
network: ers, or personal computers; they can also be devices, such as data-entry terminals.
A collection of hardware and The shared resources to which network computers can provide access include
software that enables a group files, applications, and other software; and hardware, such as printers or modems.
of computers to
communicate and provide
users with access to shared

at
resources. Basic Network Components
A network or other communication system has messages that need to be commu-

y
nicated, transmitters and receivers (senders and recipients of the message), and a
node:

op
communication channel through which the transmitter sends the message to the
A network access point.
eM
receiver. Nodes or stations are computers or other devices that communicate with
Examples include terminals
each other in a computer network. In network communication, nodes typically
and computers.
function as transceivers; they can transmit and receive messages.

message:
A chunk of data that is
transmitted over a network.
tC
No
c

transmitter:
en

The component on the

Do

″speaking″ end of a
transmission.

receiver:
se
r

The component on the

″hearing″ end of a
transmission.
fe
ea

transceiver:
Figure 1-1: Basic network components.
Pl

A device that can function as

Re

a transmitter or receiver.
Let’s look at the fundamental components of computer networks:
• Two or more nodes that desire to share information.

2
• Cabling or some other medium through which network nodes can
communicate. Typically, the network communication channel is cable; how-
ever, network communication channels can consist of other media, such as
satellite, infrared light beams, and radio.
• Software that enables network nodes to communicate. Network nodes must
use the same method or language to communicate. Network software imple-

ial
ments communication protocols. Some computer operating systems (such as
MS-DOS) require that communication software be added for network com-
munication because the systems weren’t originally designed for this type of
communication. Other computer operating systems, such as UNIX, NetWare,
Windows NT, or Windows 95/98, have built-in communication capabilities.

er
Component/Feature Definition
Server Any computer that makes its hardware and software resources available
for use by other computers on the network.
Client Any computer that uses the resources of the server.
Workstation A high-performance personal computer with specialized operating system

at
software to provide features such as multitasking, multiprocessing, and
security. Typically, workstations are used in situations where users need
to run complex and process-intensive applications on their desktop

y
computers. The term ″workstation″ is often used when ″client″ would be
more appropriate.

op
Media
eM
The physical connections between computers, such as network cabling.
Resources Files, printers, or other items accessed by clients.
Shared data Information, typically a file, provided by one computer to another across

tC
the network.
Shared printers and other Any hardware resource provided by a server to a client across a network.
peripherals
No
At the most basic level, a computer network consists of at least two computers
c

connected by a communications link, such as a physical cable. Typically, one of

the computers acts as a server, making its hardware and software resources avail-
able for use by other computers on the network. A computer that uses the
en

resources of the server is called a client .

Do
se
r
fe
ea
Pl
Re

Lesson 1: Basic Networking Knowledge 3

 ial
er
at
y
op
eM
Figure 1-2: A computer network.
The main trunk of a network communication channel is referred to as the network
backbone . Network devices can be attached directly to this backbone or be con-

tC
nected through other network devices. Cables can be connected to this backbone;
these secondary communication channels connected to the backbone are called
backbone: segments or a portion of the backbone itself.
The main trunk of a network
communication channel.
No
TASK 1A-1:
c

segment:
Identifying network components
en

Secondary communication
Do

channel connected to the

backbone. Objective: Choose the best answer for each question.

1. Which of the following correctly identifies a computer network?

a. Four computers that share a printer through a printer-sharing
device.
se
r

b. Two computers connected by a temporary serial-port connection

fe

and using file-transfer software.

ea

c. A group of computers using hardware and software to communicate

and provide users with access to shared resources.
d. Two computers, and their users, that share information through the
Pl
Re

exchange of floppy disks.

2. A server is a computer that:

a. Makes its resources available to other computers on the network.
b. Uses resources available from other computers on the network.

4
 c. Uses resources from, and makes available its resources to, other
computers on the network.

3. What are the fundamental components of a network?

ial
4. Compare the terms backbone and segment.

5. In Figure 1-3, identify the segments and the backbone.

er
at
y
op
eM
tC
No
c
en
Do

Figure 1-3: Identify the segments and the backbone.

Networking Overview
se
r

Network computing enables computers to share information and resources. How-

fe

ever, standards (rules) need to be established so that communication can occur.

ea

For example, if you’re doing volunteer work in a foreign country but don’t speak
the language and the people there don’t speak English, all of you would become
frustrated because you couldn’t understand each other. Luckily, a communication
foundation exists that includes things like interpreters and language translation
Pl
Re

books that enable people to communicate when otherwise they couldn’t. Stan-
dards make life easier for people by helping daily operations run smoothly.
Without them, chaos would result from people doing things any way they wanted,
such as driving down the middle of the street.

Lesson 1: Basic Networking Knowledge 5

 Network computing uses a similar communication foundation, which enables
computers that ″speak different languages″ to exchange data. It consists of proto-
cols , which are the means by which networking can occur. Every protocol
standardizes tasks associated with computer communication. You can think of a
protocol as a set of rules and regulations that govern a specific aspect of data
protocol: communications. A protocol might define, for example, a method for properly
A code or set of rules by beginning and ending communication and a sequence of events that should occur.

ial
which communication is
initiated, maintained, and A protocol suite is a set of related protocols that come from a single developer or
terminated. source.
• Transmission Control Protocol (TCP) is an example of a protocol.
• The Internet Protocol suite refers to a protocol suite that includes TCP and
several other related data-communication protocols.

er
Transmission Control
Protocol (TCP): A protocol stack is a set of two or more protocols that work together, with each
The component of the TCP/ protocol covering a different aspect of data communications. Typically, TCP and
IP protocol suite responsible
for establishing a connection
Internet Protocol (IP) are used in combination and might be called a protocol
between the sender and the stack.

at
receiver in a network
transmission. It guarantees
delivery of the data through
Internetworking Concepts

y
acknowledgments and
timers.
When you establish communication links between two or more networks, you

op
eM
create an internetwork. An internetwork, then, is a network of networks. You
might consider the formula for internetworking to be I=I+I (internetworking
protocol stack: equals interconnectivity plus interoperability).
A layered set of protocols

tC
working together to provide a
set of network functions.
Each protocol layer services
the layer above by using the
No
layer below.
c

Internet Protocol (IP):

en

The component of the TCP/

Do

IP protocol suite that

addresses and routes data
packets across the network
or internetwork.
se
r

internetwork:
fe

A network of networks.
ea
Pl
Re

Figure 1-4: An internetwork is a network of networks.

6
Interconnectivity
Interconnectivity refers to the ease and ability of connecting networks together.
How easily two networks connect depends, among other factors, on the type of
communication channel each network uses, and the method through which nodes
on each network access the communication channel. If both networks use the interconnectivity:
same type of communication channel and the nodes access it the same way, then The ease and ability of

ial
interconnection is relatively simple. The more disparate the systems, the more connecting networks
together.
complex interconnectivity becomes. For example, connecting two NetWare net-
works would be less complex than connecting a Macintosh AppleTalk network
and an IBM SNA network.
Internetworking devices, such as bridges and routers, connect networks or por-

er
tions of networks. These devices are discussed in greater detail later in the
course.

Interoperability

at
Interoperability refers to how communication capability takes place, and which
nodes on internetworks can communicate and share resources, such as files and

y
printers. As with interconnectivity, providing communication is a more complex
task with disparate systems than with similar systems. For example, there would interoperability:

op
How internetworks
eM
probably be a higher degree of interoperability between two IBM personal com-
puters than between an IBM personal computer and a Macintosh computer. communicate and share
resources.
Although it is an oversimplification, you might find it helpful to think of
interoperability as a software compatibility issue and interconnectivity as a hard-

tC
ware issue.

Gateways
No
c

A gateway is software or hardware (or a combination of the two) that intercon-

nects disparate types of networks, translating as necessary between the two. For
example, a gateway might connect PCs on a LAN to a mainframe. The gateway
en

is often a system dedicated specifically to the gateway task because it requires a gateway:
Do

lot of resources. Software or hardware that

connects disparate types of
Some common gateway types are described in the following table. networks, translating as
necessary between the two.
Gateway type Connects networks that use different
se
r

Protocol Protocols
Format Encoding schemes (for example, ASCII and EBCDIC)
fe

Address File systems and directory structures

ea
Pl
Re

Lesson 1: Basic Networking Knowledge 7

 TASK 1A-2:
Identifying internetworking obstacles
Objective: To identify obstacles to interconnectivity and interoperability.

ial
1. Identify factors that would make it difficult to interconnect two
networks.

er
at
2. Identify factors that would make it difficult for network computers to
share files, printers, and other resources.

y
op
eM
tC
No
c

Types of Networks
Networks come in many shapes and sizes. However, in most cases you can cat-
en

egorize them based on their geographical reach. You’ll likely see three terms
Do

describing the general size of networks: local area network (LAN), metropolitan
area network (MAN), and wide area network (WAN).
A local area network (LAN) is a collection of computers cabled together to form
a network in a small geographic area (usually within one building). A LAN can
be only as large as the physical limitations its cabling will allow. Different
se
r

LAN (Local Area cabling schemes have different limitations (which of course you can get around).
Network): However, the distinction remains—LANs are limited in size. They are typically
fe

A collection of computers inexpensive to set up, run and maintain.

cabled together to form a
ea

network in a small
geographic area (usually
within a building).
Pl
Re

8
A metropolitan area network (MAN) is slightly larger than a LAN and is con-
tained within a single city or metropolitan area. A large company might create a
MAN to link separate office buildings spread throughout a city. MANs are typi-
cally owned by private providers. They are more expensive to implement and MAN (Metropolitan Area
maintain than LANs. Network):
A network linking several
A wide area network (WAN) is relatively large and can span countries and buildings within a city or

ial
continents. Typically, WANs are privately paid for and maintained; however, com- metropolitan area.
panies often share WAN lines with other companies to save money. A company
might purchase part of the capacity of a network connection from a national pro-
vider, like MCI WorldCom. The remaining capacity of that connection will be
sold to other companies who need similar communications capabilities. Equip- Another type of network you
might see is CAN (Campus
ment owned by the network provider divides the network capacity, keeping each Area Network).

er
customer’s data separate and safe. The advantage of a WAN is that it is essen-
tially unlimited in geographic area. However, costs vary widely depending on
how the WAN is set up and configured.
WAN (Wide Area
Network):

at
A network spanning large
TASK 1A-3: areas such as multiple cities,
states, or countries.
Evaluating network basics

y
Objective: Choose the best answer for each question.

op
eM
1. Explain why the following examples are (or are not) a LAN.

A. A large office building contains 487 computers that are cabled

tC
together and share data, applications, and hardware.
No
B. A Utah-based branch office has five computers that share data and a
c

mail application with two computers in an office in Seattle.

en
Do

2. Identify the network type (LAN, MAN, or WAN) that would be most suit-
able for the needs described below. Explain why you chose that type of
network.

A. You need to exchange files and email with other employees in your
se
r

company who are located in offices throughout the United States.

fe
ea

B. You need to share network applications and printers with everyone in

your department; all of the desks are within 100 feet of each other.
Pl
Re

Lesson 1: Basic Networking Knowledge 9

 C. You need to exchange files with other departments in your company.
All departments are in different buildings throughout the city and sur-
rounding suburbs.

ial
TOPIC 1B
Network Topologies

er
A network topology refers to the layout of the transmission medium and devices
on a network. There are two types of topologies: physical and logical. Topologies
use either a point-to-point or multipoint connection scheme. A connection scheme
point-to-point connection indicates how many devices are connected to a transmission media segment or an
scheme: individual cable. An example of a point-to-point connection is a printer or modem

at
Two devices are connected connected to your computer. Another example is two computers connected
by a single communication
channel.
directly to each other to use file-transfer software such as Fastwire, MacLink
Plus, or the Microsoft Windows direct cable connection. An example of a

y
multipoint connection is a star, bus, or cellular topology network.

op
multipoint connection
eM
scheme:
Connection of three or more
Physical Topology
devices by a communication Each network has a connection scheme that describes, in general terms, the trans-

tC
channel.
mission media layout. Each network also subscribes to at least one topology,
which describes the network’s entire physical structure.
In networks that use media cable, the topology defines the cable’s actual physical
network topology:
No
configuration. Physical topologies include:
The layout of the
•
c

transmission medium and Mesh. Provides each device with a point-to-point connection to every other
devices on a network. device in the network.
• Star. Branches out via drop cables from a central hub (also called a multiport
en

repeater or concentrator) to each workstation. A signal is transmitted from a

Do

workstation up the drop cable to the hub. The hub then transmits the signal
physical topology:
to other networked workstations.
A network’s entire physical
configuration. • Ring. Connects workstations in a continuous loop. Workstations relay signals
around the loop in round-robin fashion.
•
se

Bus. Uses a single main bus cable, sometimes called a backbone, to transmit
r

data. Workstations and other network devices tap directly into the backbone
by using drop cables that are connected to the backbone.
fe
ea
Pl
Re

10
• Cellular. Refers to a geographic area, divided into cells, combining a wire-
less structure with point-to-point and multipoint design for device
attachment. The devices in each cell directly communication with a central
station called a hub.
• Hybrid. Combines two or more different topologies.

ial
er
at
y
op
eM
Figure 1-5: Physical topologies.

tC
No
Logical Topology
c

The network path that a signal follows is called the logical topology. Two logical
topologies are frequently used.
•
en

Ring. Generates and sends the signal on a one-way path, usually

Do

counterclockwise.
logical topology:
• Bus. Generates and sends the signal to all network devices. The network path through
which data travels.

TASK 1B-1:
se
r

When referred to as a logical

topology, ring and bus
Reviewing the Physical and Logical topologies
fe

topologies usually have the

word ″logical″ before them;
ea

Objective: Using the following clues, identify the appropriate topology. for example logical ring
topology.
1. Covers a geographical area.
Pl
Re

2. Uses point-to-point connection to every network device.

Lesson 1: Basic Networking Knowledge 11

 3. Workstations connect to the network in a continuous loop.

4. Uses drop cables to connect to a backbone.

ial
5. Uses drop cables to connect to a central hub.

6. Uses multiple topologies.

er
7. Generates and sends signals on a one-way path.

at
8. Generates and sends the signal to all network devices.

y
9. The structure or layout of a network’s transmission media.

op
eM
10. The path a signal follows through a network.

TOPIC 1C tC
No
c

Networking Strategies
Networks are characterized by the types of strategies they employ to interconnect
en

computers. Let’s examine the following strategies:

Do

• Types of node interaction;

• Processing strategies; and
• The services provided by networks.
se
r

Node-connecting Strategies
fe
ea

Three types of relationships exist among network nodes: hierarchical (or legacy),
client-server, and peer-to-peer.
Pl
Re

12
Hierarchical Networks
Some computer systems (typically those based on mainframes and minicomput-
ers) provide a host-to-terminal (also called primary-to-secondary) relationship hierarchical network:
between nodes. These networks are also referred to as legacy networks . Some- A network in which a host
times, you might need to add your network to an existing legacy network. This controls network
type of network can be hierarchical or centralized. In central processing, the host communications and

ial
or server performs all of the processing; users interact with the host through processing.
dumb terminals that have no processing power.
Some advantages of this type of network include centralized administration and
inexpensive terminals or nodes. Disadvantages include expensive site preparation
for the server, lack of off-the-shelf applications, processing power that diminishes

er
with each additional user login, and the expense of purchasing and upgrading
these networks.

at
y
op
eM
Figure 1-6: A hierarchical network.
tC
No
c

Client-server Networks
Servers are computers that perform a service on behalf of other network devices.
en

There are several types of servers. For example, a computer that provides other client-server network:
Do

network nodes with access to network storage devices is a file server . Print serv- A network in which duties
ers provide other network nodes with access to a printer. Both Microsoft are divided between clients,
Windows NT Server and Novell NetWare network operating systems use the which are users of network
client-server model. services, and servers, which
are providers of network
This type of network typically uses distributed processing. Applications designed services.
se
r

for client-server networks are typically stored on network drives. When a user
runs the application from a workstation, the application is loaded into the work-
fe

station’s memory, where it runs much like a non-network (local) application. This
scheme is called distributed processing because the processing of applications
ea

occurs at client stations, not in the server.

Advantages of client-server networks include processing power can be localized
to point of need, overall cost is usually much less than a hierarchical network,
Pl
Re

and servers can be optimized to provide particular services. Some disadvantages

include less centralized administration and more expensive nodes than the dumb
terminals used in legacy networks.

Lesson 1: Basic Networking Knowledge 13

 ial
er
at
y
op
eM
Figure 1-7: A client-server network.

Peer-to-peer Networks

tC
Computers that perform similar functions, both requesting and providing network
services, are peers . Networks in which no single computer controls network
communication functions are called peer-to-peer networks , or direct-cable con-
nections . Another name for this type of network is a workgroup solution .
No
peer-to-peer network: Windows 95/98 and NT Workstation include peer-to-peer support built into the
c

Each workstation on the operating system. UNIX and many of the derivatives such as Linux can operate
network acts as both a server in peer-to-peer networks.
and a client by sharing its
resources with other Advantages include that they are relatively inexpensive, processing power is
en

computers and acessing localized to the point of need (processing happens at the workstation), and all
Do

others’ shared resources. resources are potentially available over the network from every and to every
machine on the network. Some disadvantages are that there is no central adminis-
tration and that security can be difficult to control. Also, when you have more
than 15 nodes on such a network, it doesn’t work well.
se
r
fe
ea
Pl
Re

14
 ial
er
at
Figure 1-8: A peer-to-peer network.

y
Mixed Relationships

op
eM
It is possible for a network to support combinations of relationships. For
example, a network might support hierarchical, client-server, and peer-to-peer
access. In such an environment, a user might be able to access files and other
shared resources from another workstation, from a file server, or from a host.

tC
UNIX systems can be used in any or all of these relationships.
While this provides you with great flexibility and provides access to data from a
wide variety of sources, there can be many disadvantages. Administration can be
complex and the hardware and software to connect these disparate systems can be
No
expensive and complex to install. Off-the shelf solutions are often unavailable, so
c

you often need to mix and match products from several vendors, and possibly
create some of your own solutions. Some vendors won’t support their products
when you use them in such a mixed network.
en
Do

TASK 1C-1:
Identifying node interaction strategies
se
r

Objective: Identify the type of node interaction strategy that best

describes each of the following situations.
fe
ea

1. You can take advantage of several different network types.

Pl
Re

2. A mainframe computer initiates and manages network communication.

3. No single computer controls network communications.

Lesson 1: Basic Networking Knowledge 15

 4. A single computer provides specific network services to clients.

5. You can access information on the Internet.

ial
Summary

er
In this lesson, you’ve defined what a network is, identified physical and
logical topologies, and examined client-server and peer-to-peer networking
strategies.

at
LESSON 1 REVIEW

y
1A What are the three components required to have a computer network?

op
eM
tC
1B Define physical topology and logical topology.
No
c

1C What are the three relationships that exist between network nodes?
en
Do
se
r
fe
ea
Pl
Re

16
Servers and Clients LESSON
2

ial
Overview Data Files:
A network needs clients and servers in order to function. In this lesson, none
you’ll identify features of several network operating systems and the clients
Lesson Time:

er
used to access those operating systems.
1 hour

Objectives

at
To understand servers and clients, you will:

2A

y
Identify the basics of three network operating systems.
You’ll be introduced to three of the most popular network operating sys-

op
eM
tems in this topic: Microsoft Windows NT, Novell NetWare, and UNIX.

2B Identify features of various network clients.

tC
You’ll examine how to connect DOS, Windows, and Macintosh clients to
your network by using built-in and additional utilities.

2C Identify the directory services used in NT and NetWare networks.

No
You’ll be introduced to the domain model used in Windows NT Directory
Services and the Novell Directory Services used in NetWare networks.
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 17

 TOPIC 2A
Network Operating System Basics
The term network services suggests that there is an entity that requests a service

ial
and another entity that provides the service. A service provider consists of a com-
puter as well as the software enabling the computer to fulfill requests. The
network operating system (NOS) contains the applications that provide network
services. While the local operating system controls the resources for the computer
on which it resides, the network operating system manages the network resources
so that they can be adequately shared among the network users.

er
Let’s examine the role of servers, clients, and peers in networks. Servers are gen-
erally service providers, clients are generally service requesters, and peers can be
both. The active software determines whether a computer is a server, client, or
network operating peer. It is possible for one computer to act as a server, client, and a peer at the
system: same time; however, most computers fulfill only one role at a time.

at
A collection of software that
enables network nodes to To work effectively, individuals need access to data, processing, and input/output
communicate with other resources. Through the network operating system, users gain access to network

y
nodes. services. Granting this access to users is transparent.
Network services include file, message, print, security, routing, network adminis-

op
eM
tration, application, and database. These are common to all major network
operating systems.
Let’s examine three different network operating systems. These are Windows NT,

tC
Novell NetWare, and UNIX. They are the three most popular network operating
systems.
No
Windows NT
c

Windows NT is a 32-bit operating system that can run on Intel (and compatible)
and RISC platforms. It is a preemptive multitasking operating system that shares
en

processor time between all running processes. NT is compliant with the govern-
Do

ment’s C2 security specifications. It can be used to build a network, intranet, or

Internet server. Win32 API, 16-bit Windows, and MS-DOS based applications are
supported on the NT platform. NT comes with the Internet Explorer Web browser
and the email software Microsoft Exchange Client.
se
r
fe
ea
Pl
Re

18
 ial
er
at
y
op
Figure 2-1: The Microsoft networking family.
eM
The following table describes four different editions of Windows NT.

tC
Windows NT edition Designed for
Workstation Desktop operating system for business applications.
Server General network operating system for any size network.
No
Enterprise Server Large, mission-critical applications.
c

Terminal Server Non-PC devices, such as dumb terminals, to be part of an NT

network via terminal emulation.
en
Do

The BackOffice runs on NT Server. BackOffice includes:

BackOffice component Is used for

Exchange Server Messaging and collaboration services
Proxy Server Firewall and Web-caching services
se
r

Internet Information Server (IIS) Internet and intranet services

Site Server Intranet, Internet, and e-commerce management services
fe

Systems Management Services Inventory, network diagnostics, and software distribution services
ea

(SMS)
SNA Server Integrating legacy systems with an NT network
SQL Server Relational database management
Pl
Re

NT Server is a network operating system. It includes Gateway Service for inte-

grating NetWare and NT networks and NetWare Migration utility to migrate
NetWare users and data from NetWare servers to NT servers. You can also pur-
chase additional products that enable you to administer both servers from a single
location.

Lesson 2: Servers and Clients 19

 Fault tolerance features of NT Server include disk mirroring, disk duplexing, and
disk striping with parity.
For Internet support, NT Server includes TCP/IP support, Domain Name System
(DNS) server support, IIS, and FrontPage (HTML page editor) software.

ial
er
at
y
op
eM
Figure 2-2: A domain.
tC
NT Server uses a domain model. A domain is a logical grouping of computers
No
and devices on a network administered as a single entity by using a common set
c

of rules and procedures. There can be one or more domains per network.
domain: Domains can be configured to share resources with other domains, but also might
An organizational unit on a not be configured to share resources. Systems identified as being a member of a
en

Windows NT network in domain might be primary and backup domain controllers, member servers, or
Do

which workstations and

servers share a single
client workstations. Users who log in to the domain have access to all servers in
accounts database and are the domain.
administered as a group. NT Servers are generally installed as primary or backup domain controllers. In
order to change domain membership you’d have to reinstall NT.
se
r

Only one primary domain controller exists per domain. It contains the domain
accounts database and the user and group database. It is recommended that at
least one backup domain controller be installed per domain. It contains a copy of
fe

the domain accounts database. Changes made to the primary domain controller
ea

copy of the database are replicated to all backup domain controllers in the
domain. Backup domain controllers can be used to validate user accounts and can
be promoted to a primary domain controller if something happens to the original
Pl
Re

primary domain controller. Geographically dispersed domains should have at least

one backup domain controller at each location so that logons can be handled
locally, reducing traffic over the WAN link.

20
Comparing Domains and Workgroups
NT can also use a workgroup model rather than domains. A workgroup is used on
a peer-to-peer network. It’s typically used in a very small network that doesn’t
require as secure of an environment as that provided by domains. Every computer
in the workgroup maintains its own account information, and doesn’t share it
with other systems. Resources are accessed by knowing and entering the pass-

ial
word assigned to the resource.

er
at
y
op
eM
Figure 2-3: A workgroup.
tC
No
c

NT Superuser
en

Network operating systems need to have a user with full access and rights to
Do

everything on the network in order to administer it. This user is referred to as the
superuser. Different NOSes have different superuser account names.
You can also define additional users with the same rights and access as the
default superuser. This is useful in case you forget the password of the superuser superuser:
se

or if you need to share administration responsibilities between several people.

r

A network account with full

access and rights to
The superuser on an NT system is the user Administrator. This account has com- everything on the network in
fe

plete, unlimited access over the operating system, whether it is used as a order to administer it.
ea

workstation or a network server.

Pl
Re

Administrator:
The superuser account on an
NT system.

Lesson 2: Servers and Clients 21

 NetWare
Novell’s NetWare product is another popular PC network operating system.
Workstations use Novell Client software to log in to and communicate with the
network. NetWare workstation software also enables workstations to determine
Novell was founded in 1983. the operations to be performed by the workstation and the ones to be delegated to
the network. Servers use NetWare to provide and control workstations’ access to

ial
network resources, including shared printers, storage, and communications
devices. NetWare 5.x is the most recent version of the operating system available.
As a network operating system, NetWare provides services to network users. The
following table describes some of these services.

er
Type of service Description
File-storage services Enable users to store files on and retrieve files from the server’s
hard disk.
Novell Directory Services (NDS) Provide administrators and users with a way to manage and
access network resources.

at
Security services Ensure that network data and resources are available only to those
users who should have access to them. NetWare 5.0 has the
government’s C2 security rating.

y
Print services Enable users to share printers that are connected to the network.

op
Application services Enable administrators to set up and configure network-based
eM applications so that network users have easy access to them.
Storage Management Services Enable administrators to back up and restore data located on both
(SMS) servers and workstations. Data is usually backed up to a tape

tC
storage device.
Connectivity services Enable administrators to ensure Internet and remote access to
users who require such access.
Web services Enable administrators to integrate Web servers and solutions on
No
the network.
c

Database services Enable administrators to deliver database solutions on the

network.
Communications services Enables use of third-party communications and messaging
en

systems including GroupWise, cc:Mail, MHS, and SMTP.

Do

Multiprotocol Routing (MPR) Enables you to connect your network to public and private
networks using services such as T1, frame relay, or ISDN.
Symmetric Multiprocessing (SMP) Enables use of SMP utilities on the network.
se
r
fe
ea
Pl
Re

22
NetWare versions 4.0 and higher use a directory service based network, Novell
Directory Services or NDS. NDS maintains information about all network
resources (such as servers, volumes, users, groups, and printers) in a hierarchical,
tree-structured database that resembles an organizational chart. With it, system
administrators can organize network resources according to the way a given NDS:
workgroup is structured—an efficient method for giving users access to the net- Novell Directory Services.
work according to the role they play in the organization. When a user requests a Formerly known as NetWare

ial
Directory Services. An
resource, NDS locates the resource, verifies that the client is valid, and makes the administrative database that
resource available to the client. defines network entities,
NetWare 3.x and previous versions used a bindery-based database to maintain such as users, groups,
servers, and so on. NDS was
information about users and groups. This wasn’t an enterprise-wide database; a introduced in NetWare 4.0,
separate database was required on each server in the network. This means that if and controls access to

er
you need access to three servers, you’ll need three separate accounts (one on each network information,
server). resources, and services. It
allows users to access all
The default protocol for NetWare 5.0 is IP; previous versions used IPX/SPX as network resources with a
the default protocol. Both protocols are supported. single login procedure.

at
NetWare 5.0 includes several additional products with the operating system.
These include a 5-user version of Oracle, Netscape FastTrack Server, FTP Server,
IPX/IP Gateway, DHCP Server, DNS Server and Client, Netscape Navigator, and The default protocol for

y
others. NetWare 4.x and earlier is
IPX/SPX.

op
eM
NetWare Superusers
The superuser on a NetWare system is the user Admin. This account has com-

tC
plete, unlimited access over the operating system. This user is added when the
first server in the tree is installed. You can add other users with full administra-
tive rights, but this is the default superuser for NDS-based NetWare networks. admin:
NetWare 3.x and previous versions used SUPERVISOR rather than Admin, and a The superuser on NetWare
4.x and higher networks.
No
separate SUPERVISOR account was required for each server on the network.
This user has full
c

administrative rights to the

Directory tree objects and to
UNIX the file system.
en
Do

UNIX is a family of operating systems developed at Bell Laboratories in the late

1960s. Because UNIX was designed for software development, the needs of the
beginning user were ignored in favor of speed and accuracy. However, UNIX has SUPERVISOR:
evolved with an ever-increasing focus on the end-user’s needs. Today, many non- The superuser account on
NetWare 3.x and lower
programmers use UNIX effectively in a variety of ways.
networks. This user has full
se
r

administrative rights to
create users, groups, print
services, and to administer
UNIX Features and Benefits
fe

the file system.

ea

The many features and benefits of UNIX include portability, portable applications
software, multiuser operations, multitasking, utilities, pipes, and software develop-
ment tools.
Pl
Re

One of the main reasons for the longevity and popularity of UNIX is its
portability. UNIX runs on almost every type of computer from micros to super
computers. Unlike most other operating systems, UNIX is almost entirely written
in C, a high-level programming language. Therefore, moving UNIX to a new sys-
tem can be done in a relatively short amount of time.

Lesson 2: Servers and Clients 23

 The applications software written for UNIX is also portable. With only minor
modifications, the same UNIX-based applications can run on any computer that
runs UNIX.
UNIX can function as a workstation and as a network operating system. It is
often configured as a server, but can also be part of a peer-to-peer network, or it
can be used as a workstation on a UNIX or other type of client-server or hierar-

ial
chical network.
This multiuser, multitasking operating system can perform multiple tasks for each
user simultaneously. One of the initial objectives of UNIX was to run the same
operating system on different computers and to enable data communications
between them. Consequently, UNIX was designed with networking in mind.
Many companies have changed to UNIX because of its networking capabilities.

er
There are many built-in and third-party applications and utilities that enable
UNIX to be connected to other networks, such as VMS, MVS, NetWare, and NT.
UNIX provides utilities for everything from email to compilers. There are over
300 utility programs and commands in the UNIX environment. The commands

at
are case-sensitive and the command names can be somewhat cryptic. You can use
pipes, one of the most famous UNIX features, to combine several simple utilities
to perform more complex functions.

y
There are many versions of UNIX available. Some of the major UNIX vendors
include Pyramid, HP, IBM, and Sequent.

op
eM
There are also a lot of UNIX-compatible operating systems, such as Linux. Linux
is a POSIX based operating system that can run on many hardware platforms
including PC, Macintosh, Sparc, and Alpha systems. Some vendors that supply

tC
Linux was developed by Linux are Red Hat, Caldera, Suse, and Debian.
Linus Torvolds at the
University of Finland. It is a
freely available operating
system with open source
No
code.
c

POSIX:
en

Portable Operating System

Do

Interface for UNIX. A set of

standards developed by IEEE,
specifying components of a
portable, vendor-independent
operating system. The goal
of POSIX is to enable the
se
r

development of portable
application software across
fe

different vendors’ hardware.

ea
Pl
Re

24
UNIX Superuser
The superuser on a UNIX system is the user root. This account has complete,
unlimited access over the operating system, whether it is used as a workstation or
a network server.
root:
The superuser on UNIX

ial
systems. This is a special
Cross-platform Communication login name used by the
system administrator to gain
Since many networks today combine two or more different network operating full administrative access to
systems into a single network, you might need a solution in which all servers, as the system. Logging in as
well as clients, can communicate with one another. Some of the solutions in Win- root enables you to perform
dows NT Server that allow NT clients to access NetWare servers include any valid command on any

er
Gateway Service for NetWare, Client Service for NetWare, File and Print Ser- file, regardless of what the
permissions on that file are.
vices for NetWare, and Directory Service Manager for NetWare. NT Server
includes UNIX printing support services.
Novell has a product, NDS for NT, which allows NT users to access both NT and

at
NetWare through NDS with a single login. Novell NOSes also include File and
Print Services for UNIX-based clients. You can also purchase products such as
NetWare for SAA for connecting to IBM SAA mainframes.

y
op
TASK 2A-1:
eM
Identifying features of three popular network

tC
operating systems
1. What are the three major network operating systems?
No
c

2. Who is the administrative user for each of the three major operating
systems?
en
Do

3. What are the core services provided by the network operating systems?
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 25

 TOPIC 2B
Network Clients
Clients are the portion of the network through which users access the network.

ial
They provide the connectivity between the server and the user. The choice of
which client you use can greatly affect the network performance in a client-server
network. Clients can include MS-DOS, Windows (in all of its many varieties
from Windows 3.x to Windows NT Workstations), Macintosh, and NetWare
clients.
Whenever possible, you should choose a client whose operating system is well-

er
matched to the network operating system you’re using. It should be able to run
the protocols available on your server.
You might need to install network-specific software or redirection software on
clients in order for them to communicate with the network. Your client might

at
need to run more than one piece of client software if it needs to connect to vari-
ous network operating systems simultaneously.

y
MS-DOS Clients

op
eM
DOS, by itself, doesn’t include any network connectivity features. Client software
that is DOS-compatible must be installed in order for a DOS-based client to
access the network. The client software that you install will depend on the NOS

tC
to which you are connecting.
If you are using a real-mode client, only one client can be loaded at a time. This
can be a problem if users need to connect to two different network operating sys-
tems, each requiring different client software.
No
real mode:
c

Program instructions can

access any address space in
the first MB of RAM. Usually Windows 3.x and Windows for Workgroups 3.x Clients
en

these programs run in real

While Windows 3.x is network aware, it is not really an operating system since it
Do

mode because they need to

access and use system requires DOS in order to run; therefore, some of the limitations of the DOS client
information. Compare with also apply to these clients.
protected mode.
Many external clients are available for Windows 3.x even though none came with
the software. For most client software, SHARE.EXE should be loaded before
se

loading Windows to prevent multiple applications from corrupting a data file by

r

SHARE doesn’t need to be trying to simultaneously access it.

loaded when using the
fe

Microsoft Network Client Only one network client can be active at once since Windows 3.x uses only real
ea

software. mode network client software. Some examples of the client software available
include:
• Microsoft Network Client
Pl

•
Re

Novell NetWare
• Artisoft LANtastic
• Banyan VINES

26
Each of these clients is used to attach to its own network (for example, NetWare
client attaches the client to a NetWare network, and LANtastic attaches the client
to a LANtastic network).
Windows clients (from
Windows for Workgroups 3.x allows for the use of protected mode software, Windows 3.x through
enabling you to have more than one client loaded at once. It is also a peer-to-peer Windows NT) can all use IPX
networking version of Windows 3.x that enables any system to function as a cli- (called NWLink in Microsoft

ial
ent or a server on the network. terms), TCP/IP, and NetBEUI
protocols to communicate
across the network.

Windows 95/98 Clients

Windows 95 and Windows 98 support both real-mode and protected-mode client protected mode:

er
software. The client software is installed, configured, managed, or removed Program instructions can
through the Network Control Panel. You can only install and use one real-mode access only specific
client at a time. However, you can use as many protected-mode clients at the contiguous areas of the first
same time as you need to use. 640 KB of RAM. The local
operating system typically
Windows 95 real-mode client software includes Banyan VINES DOS/Windows runs in protected mode to

at
3.1, FTP Software NFS client, Novell NetWare Workstation Shell 3.x (NETX), prevent data from being
Novell NetWare Workstation Shell 4.0 and above (VLM), and SunSoft PC-NFS overwritten by other
programs. Compare with real
(5.0). The protected-mode client software includes Microsoft’s Client for NetWare

y
mode.
Networks and Client for Microsoft Networks. The protected-mode clients use no
conventional memory.

op
eM
Windows 98 also includes the Microsoft Family Logon protected-mode client
software. In addition, third-party protected-mode client software is available for
Artisoft LANtastic 7.0 or above, Banyan 32-bit or 16-bit, Digital PATHWORKS

tC
32, and IBM Networks Client for Windows 95.
Novell also has a network client that you can install, called Novell Client (previ-
ously known as Client 32). It is a protected-mode application available for
accessing all versions of NetWare from 2.x through 5.x by using Windows or
No
DOS clients. You can try both Novell Client and the Microsoft Client for
c

NetWare networks to determine which one gives you the best performance on
your network.
en
Do

Macintosh Clients
Macintosh computers come preinstalled with AppleTalk network software. This
enables Macs to be quickly networked to one another; however, if you want to
add your Mac to a NetWare or NT network, you’ll need to install support on the
se
r

client, the server, or both.

NT Server includes Mac client support, but the Services for Macintosh need to be
fe

installed on the server to activate this support. This provides file and printer shar-
ea

ing, and allows you to administer Macintosh users the same way PC users are
administered.
NetWare 4.1x networks are automatically configured with the Macintosh name
Pl
Re

space on the SYS volume to support Macintosh file formats. On other NetWare
versions of NetWare, you will need to load the MAC name space. Also, you need
to install the NetWare Client for Macintosh on each Mac that will be accessing NetWare 5 doesn’t include
the NetWare server. This adds features to the Mac including a login menu on the Macintosh connectivity
support; Prosoft Engineering
menu bar, full file and print resource access, NetWare Volume Mounter utility for
now provides it.
fast NetWare volume access, MacIPX so that Macs can use IPX/SPX, NetWare/IP
support, and support for the built-in Ethernet port on Power Macs.

Lesson 2: Servers and Clients 27

 Linux Clients
Linux workstations are also becoming popular client platforms. Linux supports
TCP/IP, IPX, and NetBEUI protocols, so can communicate with many different
network operating systems.

ial
TASK 2B-1:
Identifying client features
1. Identify which systems in your classroom are clients.

er
2. Your network includes Windows NT 4.0, NetWare 4.11, and Linux 6.0

at
servers. What software would need to be installed on your client in
order to communicate with the network? Is that software included with
the client operating system or does it come from the network operating

y
system?

op
eM
3. What other clients could have been included on this network? What

tC
software would be required for the clients to communicate with the
classroom servers?
No
c
en
Do

Suggested time:
APPLY YOUR KNOWLEDGE 2-1
10 minutes
Logging on to the network and accessing resources
se
r

In this activity, you’ll log on to the network as the NT and NetWare superuser
accounts, then run applications from the servers.
fe
ea

1. If your computer is already at the Windows 95 desktop, choose Start→

Shutdown→Close All Programs And Log On As A Different User.

2. When prompted, log in to the NetWare server as Admin, using the password
Pl
Re

manage. This is the NetWare superuser.

3. When prompted, log on to the NT server as Administrator using the pass-

word password. This is the NT superuser.

28
4. Right-click on Network Neighborhood and choose Map Network Drive. In
the Path text box, enter \\NTSRVR\C$ . This is the default share volume on
an NT Server. Click OK.

5. Open Explorer. Examine the list of drives. The local drives are listed, then
the drive on the NT server you just mapped, followed by the default drive
mappings from the NetWare login.

ial
6. Expand C$ on NTSRVR, Program Files, Windows NT, Accessories, and
open the ImageVue folder. Double-click on WANGIMG to start the Imaging
For NT program from the NT server. You are now running this program
from the server. Choose File→Exit to close the application.

er
7. Expand SYS on 411SRVR, Public, and then open the Win95 folder. Double-
click on NWADMN95 to start the NetWare Administrator utility. You are
now running this program from the NetWare server. Choose Object→Exit to
close NetWare Administrator.

at
TOPIC 2C

y
Directory Services

op
eM
Today, many organizations seek to connect all of their computing resources in a
single, enterprise-wide network—an objective that greatly increases the complex-
ity of system administration. In an enterprise that includes thousands of network

tC
users, even seemingly straightforward tasks (such as creating user names and
authorizing access to network resources) can be troublesome.
Directory services help alleviate some of the problems inherent in administering
large networks. Its main purpose is to regulate access to network information,
No
resources, and services. It is very much like using an operator or the Yellow
c

Pages to find a telephone number or information about a company (or person).

When you are designing your network, you should keep two important goals in
en

mind:
Do

• Each user should be able to access all network resources with a single
account and password.
• You should be able to centrally manage the network from a single computer.
se
r

Windows NT Directory Services

fe

By using Windows NT Directory Services or NTDS, you can create a user only
ea

once in the accounts database, yet you can give this user access to all network
resources. NTDS integrates with Microsoft BackOffice and NetWare; thus, all
users can access these resources through their Windows NT accounts.
Pl
Re

Using NTDS simplifies network management in two ways. NTDS supports cen-
tralized management of the accounts database, no matter how large, from any
computer on your network. NTDS also enables you to define a user’s account
once for your entire network.

Lesson 2: Servers and Clients 29

 When designing your network, you implement NTDS through the use of domains
and trust relationships. You create a Windows NT domain when you group com-
puters and users together for administrative purposes. You add users to the
domain by creating them within User Manager for Domains. You add computers
to the domain either by adding them in Server Manager on a domain controller,
or at the time they ″join″ the domain through their own Control Panel in Win-
dows NT.

ial
Your Windows NT network can consist of one or more domains. Because
domains are used to logically group users and computers, you may find that you
need more than one domain to manage your network. Some of the factors that
influence the domain structure for your network include:
• Geographical —where you establish domains at each geographical location

er
of your company.
• Functional —where you define domains following the company’s organiza-
tional chart (usually departmental).
• Administrative —where you create domains based on who will be managing

at
them.
Once you have designed your domain structure, you might face a dilemma if you
have created multiple domains: what if users need to access resources outside

y
their own domains? Windows NT Directory Services solves this dilemma through
The example in the text is of the use of trust relationships. Through a trust relationship, you can establish a

op
two one-way trusts between
eM
secure connection between two domains. After you create the trust, you can then
domains. The trust
give the users from both domains access to either domain’s local resources.
relationship could also be
one-way only. The goal of Windows NT Directory Services is for a user to have only one

tC
account and password, yet have access to all network resources, provided that the
user has been given the appropriate permissions. Although your network may
consist of multiple domains, the ability to establish trust relationships between
those domains is what enables users to have only one account yet still have
No
access to all network resources.
c

A user can log on to your network from any computer in a trusting domain by
using pass-through authentication. Pass-through authentication automatically for-
wards a user’s logon request to the appropriate domain for logon verification.
en
Do

Network Administration
After you have established trust relationships between your domains, you can
manage your network from any computer. The following table identifies some
se
r

common management tasks and the utilities you use to perform them.
fe

Utility Manages
ea

User Manager for Domains User and group accounts

Server Manager Domains and computers
Pl
Re

Both of these utilities enable you to manage all users and resources in all
domains on your network. To manage a different domain, simply select the appro-
priate domain within the utility.

30
You can create users and then use groups to help administer multiple users rather
than assigning resources and rights individually to each user, you can make
assignments to groups. Local groups contain users within a single domain. Global
groups are used to organize groups across multiple domains.

local group:
Groups of users in a single

ial
TASK 2C-1: NT domain.

INSTRUCTOR DEMONSTRATION—Creating users and

adding users to groups in NT global group:
Groups of users across
Objective: To use User Manager for Domains to create users and groups. multiple NT domains.

er
Setup: Observe as your instructor performs the steps in this task.

1. If necessary, log in as Administrator to the NTSRVR in the NETDOMAIN.

at
2. Choose Start→Programs→Administrative Tools (Common)→User Man-
ager for Domains.

y
3. Choose User→New User. In the New User dialog box, enter a username,
full name, and password. Uncheck User Must Change Password At Next

op
Logon. Click Add.
eM
4. Add another user (or more), and then close the New User dialog box.

tC
5. Next, let’s create a global group. Choose User→New Global Group. Name
the group GL-Students.

6. In the Not Members list box, select the student user names (Stu#) and
then click Add. Click OK.
No
c

7. Now, let’s create a local group. Choose User→New Local Group. Name
the group LCL-Group.
en

8. Click Add to display the Add Users And Groups dialog box. Select the user
Do

names created in this task, and then click Add. Notice that the Add
Names box contains the names in the format domain-name\user-name sepa-
rated by semicolons. Click OK twice.
se
r

Understanding Trust Relationships

In the Windows NT environment, you use trust relationships to establish secure
fe

links between domains. Trust relationships are configured through User Manager
ea

for Domains. When one domain trusts another, that domain’s users and global
groups are available from within User Manager for Domains and when configur-
ing share permissions. Once you have established trust relationships, you can give
Pl

users access to resources outside their domains. You can also manage all trusted
Re

domains from any computer.

Lesson 2: Servers and Clients 31

 Trusted and Trusting Domains
A trust relationship is a secure communications link between two domains so that
users from one domain may access resources in the other domain.
A trusted domain is a domain that is part of a trust relationship with another
domain. When a domain is trusted, the trusted domain’s users can be granted per-

ial
missions to other domains’ resources.
A trusting domain is a domain that makes its resources available to users from
other domains.
trusted domain:
A domain that is part of a
trust relationship with
Types of Trusts

er
another domain. When a
domain is trusted, the trusted
domain’s users can be The trust relationships between domains can be either one-way or two-way. In a
granted permissions to other one-way trust relationship, one domain trusts the other domain. In a two-way
domains’ resources.
trust relationship, both domains trust each other.

at
A one-way trust relationship is a trust relationship between two domains where
only one domain trusts the other domain. In a one-way trust relationship, the
trusting domain: trusted domain can access resources in both domains.

y
A domain that is part of a
trust relationship with A two-way trust relationship is a trust relationship between two domains where
both domains trust each other. In a two-way trust relationship, users from both

op
another domain. When a
domain is trusting, it makes
eM
domains can access resources in either domain.
its resources available to
users from other domains.

tC
Guidelines for Designing Trust Relationships
one-way trust When you design the trust relationships for your network, you should keep the
relationship: following guidelines in mind:
•
No
A trust relationship between You can only create trust relationships between Windows NT domains.
two domains where only one
c

domain trusts the other • You should establish the least number of trusts necessary for your environ-
domain. In a one-way trust ment to reduce the administrative overhead of managing those trusts.
relationship, the trusted
• The physical location of users does not matter since pass-through authentica-
en

domain can access resources

tion enables users to log on to the network across trusts.
Do

in both domains.
• Trusts are not inherited from domain to domain; they are non-transitive.
Consider the following example where the San Francisco domain trusts the
Los Angeles domain and the Los Angeles domain trusts the San Diego
two-way trust
domain. The San Francisco domain does not inherit a trust relationship with
relationship:
se

the San Diego domain from the Los Angles domain.

r

A trust relationship between

two domains where both
domains trust each other. In
fe

a two-way trust relationship,

ea

users from both domains can

access resources in either
domain.
Pl
Re

32
 ial
er
Figure 2-4: Trusts relationships are non-transitive.

at
Establishing Trust Relationships

y
When establishing a trust relationship, you must configure both domains from

op
eM
within User Manager for Domains to make the trust successful. You should con-
figure the trusted domain first, then configure the trusting domain. Although you
can initiate the trust relationship from either domain, you will receive an error
message if you start with the trusting domain because it will be unable to verify

tC
the trust. You will be able to use the trust immediately if you configure the
trusted domain first. However, if you configure the trusting domain first, it may
take up to 15 minutes for the trust relationship to be established.
No
c

TASK 2C-2:
en

Examining Windows NT Directory Services

Do

considerations in your network design

Objective: Read the following scenario, and then answer the questions
posed in the steps below. You’re planning to implement a net-
work consisting of 10 Windows NT servers, 5 NetWare
se
r

servers, a Microsoft SQL server with an important customer

database, and 2500 clients. You’d like to minimize administra-
fe

tive overhead.
ea

1. Define the following terms:

domain:
Pl
Re

trust relationships:

Lesson 2: Servers and Clients 33

 pass-through authentication:

2. With respect to management of user accounts, passwords, and security,

ial
what are some of the goals you might have?

er
3. How might you design the domain structure and, if necessary, the trust

at
relationships for this network?

y
op
eM
tC
Suggested time:
CHECK YOUR SKILLS 2-1
Under 5 minutes
Considering another scenario
No
c

Your network consists of two sites, one in San Francisco and the other in Los
Angeles. You have configured each site as a separate Windows NT domain. Your
user account has been created in the San Francisco domain.
en

How can you configure the network so that all users can log on from any
Do

computer at either site?

se
r

Windows NT Domain Models

fe

There are four different strategies you can use to design your domains and trust
ea

relationships. These strategies are referred to as domain models and include

single domain, single master domain, multiple master domain, and complete trust
domain.
Pl
Re

34
The Single Domain Model
In the single domain model, your network consists of only one Windows NT
domain. Thus, all user accounts and resources are contained within the same
domain. The single domain model supports the Windows NT Directory Services
goals of one account per user with access to all network resources and centralized
administration. Because the network consists of only one domain, you don’t need

ial
to configure any trust relationships.
When using this model, your network should have one Primary Domain Control-
ler (PDC) and at least one Backup Domain Controller (BDC). A single domain
can contain approximately 40,000 accounts in the directory services database.
Advantages of using this model include centralized administration of accounts

er
and resources and that it is the easiest model to implement because no trust rela-
tionships are required. Disadvantages of using this model include poor
performance if the domain is large and the need for powerful hardware for
domain controllers to maintain performance. Also, resources can’t be grouped as
they would be if they were in separate domains.

at
The Single Master Domain Model

y
The single master domain model consists of multiple domains. The master

op
eM
domain contains all user accounts for the network, and is therefore similar to a
single domain. The main difference between the two models is that the single
master domain model also contains resource domains. You must establish one-
way trust relationships between the master domain and the resource domains so

tC
that the user accounts can access any of the resources. After the trust relation-
ships are established, you can then give any user account access to resources in
any of the resource domains.
No
Each domain in your network should have a PDC and at least one BDC.
c
Although the model supports multiple domains, all user accounts are contained
within the master domain. Thus, you’re still limited to a maximum of 40,000
users in the directory services database.
en

Advantages of this model include centralized administration of user accounts and

Do

decentralized administration of network resources. Resources can be grouped into

domains either by location or by department.
Disadvantages of this model include poor performance if the domains are large,
each resource must have its own local groups for granting permissions to
resources, and that only the administrators of the master domain can create users
se
r

and global groups.

fe
ea

The Multiple Master Domain Model

If your network requires more than 40,000 user accounts, then one design strat-
egy you can use is the multiple master domain model. In this model, you
Pl
Re

configure two or more master domains and multiple resource domains. Together
the master domains contain all of the user accounts for the entire network. The
resource domains contain the network resources and are usually defined based on
location or department.

Lesson 2: Servers and Clients 35

 To maintain the goal of a single user account for each user for the entire network,
the master and resource domains must be able to exchange user information
through trust relationships. Each master domain must have a two-way trust rela-
tionship with all other master domains and each resource domain must have a
one-way trust relationship with all master domains; the resource domain must
trust the master domains.

ial
Advantages of this model include:
• Accommodation of more than 40,000 users.
• Centralized administration of user accounts.
• Decentralized administration of network resources.
• Resources can be grouped into domains either by location or department.

er
Disadvantages of this model include:
• It is more complicated to design and manage because of the large number of
trust relationships.
• User accounts are split over multiple domains.

at
• It must create global groups in each of the master domains.

y
The Complete Trust Domain Model

op
eM
In the complete trust domain model, all domains on your network are connected
via two-way trust relationships. This model is similar to the multiple master
domain model, in that it supports networks with more than 40,000 users. How-

tC
ever, the administration of a complete trust network is completely decentralized.
Advantages of this model include:
• It can accommodate more than 40,000 users.
No
• It accommodates companies that don’t have a central network management
c

staff because the administration of network resources and user accounts is

decentralized.
• It can group resources into domains either by location or by department.
en

•
Do

Each domain has complete control over its user accounts and resources.
Disadvantages of this model include:
• It is more complicated to design and manage because of the large number of
trust relationships.
•
se

It doesn’t accommodate companies with a central network management staff

r

because the administration is decentralized.

•
fe

It may have security problems if domain administrators put inappropriate

ea

global groups into their local groups.

Pl
Re

36
TASK 2C-3:
Discussing domain models
1. What are some benefits to implementing the single domain model?

ial
2. How does administering a single master domain model differ from
administering a single domain model?

er
at
3. Why would you choose to implement the multiple master domain model
instead of the single master domain model?

y
op
4.
eM
How should you determine where to create your users’ accounts in the
multiple master domain model?

tC
No
c

5. When using the multiple master domain model, where should you create
the necessary global groups?
en
Do

6. Why would you choose to implement the complete trust domain model?
se
r
fe
ea

7. What is a risk of implementing the complete trust domain model?

Pl
Re

Lesson 2: Servers and Clients 37

 Novell Directory Services
Novell Directory Services (NDS) is a major feature of NetWare, starting with
NetWare 4.0 and continuing through the current versions of 4.x and 5.x. Previous
versions of NetWare were bindery-based (a flat-file database structure).
NDS (Novell Directory
Services): NDS maintains information about all network resources (such as servers, vol-

ial
Formerly known as NetWare umes, users, groups, and printers) in a hierarchical, tree-structured database that
Directory Services. An resembles an organizational chart. With it, system administrators can organize
administrative database on network resources according to the way a given work group is structured—an
the network that defines efficient method for giving users access to the network according to the role they
network entities, such as play in the organization. When a user requests a resource, NDS locates the
users, groups, servers, and
so on. NDS was introduced
resource, verifies that the client is authorized to access the resource, and makes

er
with NetWare 4.0, and the resource available to the client.
controls access to network
information, resources, and
services. It allows users to
access all network resources Examining the NDS Database and its Components
with a single login

at
procedure. NDS enables users to access network information and resources. Every resource
on the network has an entry in the NDS global database. This database, called the
Novell Directory, represents network resources as database objects that can be

y
organized independently of physical location. All NetWare 4.x and 5.0 servers
connected to a single network have access to, and use, the same Novell Directory.

op
eM
Therefore, they have information on all available resources on the network. Thus,
a single point of access exists for both using and managing network resources.
When you want to access a resource, it is not necessary for you to know the
name of the network server providing that resource. To gain access or receive

tC
The NDS tree is also called information about the resource, you simply identify it by name, provided that you
the Directory tree. You might have the proper rights to it.
encounter the term Directory
tree in NetWare-related
documents.
No
The NDS Tree
c

Because NDS is critical to network management and security, you must carefully
plan the NDS structure. The Directory objects are organized in a hierarchical
structure called the NDS tree ( or Directory tree ). The NDS tree is similar to the
en
Do

tree structure found in DOS file systems.

se
r
fe
ea
Pl
Re

38
Like the DOS root directory, the [Root] object exists at the top of the tree. Con-
tainer objects, analogous to DOS directories, can exist within either the [Root]
object or each other. Leaf objects, the structural equivalent of DOS files, can be
placed within Container objects. [Root] object:
Represents the top of the
The NDS tree can contain an unlimited number of container levels, so you can NDS tree and is created
set up a logical structure that reflects your organization’s network environment. during the installation of the

ial
However, you should try to keep the number of levels in your tree to a minimum; NetWare network operating
otherwise, they might become unwieldy to administer. NetWare documentation system.
indicates that you should not have more than six levels in your NDS tree.
The Novell Client software enables you to browse the NDS tree using Windows
Explorer and Network Neighborhood. It is the only client software for Windows Container object:
95 and Windows NT Workstation that provides full support of NDS. An object that can contain

er
other objects and is used to
Note: The NDS tree, unlike a DOS directory tree, has restrictions on where specific organize the Directory tree.
Container objects can be placed.

at
Leaf object:
An object that can’t contain
other objects, such as a User
object.

y
op
eM
tC
No
c
en
Do
se
r

Figure 2-5: NDS Directory structure.

fe

You can set up an NDS tree in different configurations. Some examples include:
ea

• Shared resources. If personnel from different departments need to access a

particular printer or server, you can arrange the NDS tree to provide access.
Consider carefully how you
• Job levels. Vice-presidents, division managers, and department managers can design the NDS tree. The
Pl

be grouped together along with their respective resources. stability of your company
Re

organizational chart and floor

• Plant location. If the company has plants in several locations (on streets plan should also be taken
such as Corporate Boulevard, Winding Way, and Elm Street), you could into consideration when
organize the NDS tree by using Container objects named CorpBlvd, evaluating examples such as
WindingWay, and Elm. those in this list.

Lesson 2: Servers and Clients 39

 • Floor. If departments located on the same floor in the company share many
of the same resources, you might organize the NDS tree by floor or by part
of the building.

Tree Design

ial
You should strive for a tree design that is easy to administer and that provides
users with access to the resources they need. NDS trees can be based on geo-
graphic location, organizational structure, administrative responsibilities, or any
combination of these.
This flexibility in tree design also brings with it added complexity, so careful

er
planning before you implement your NDS tree can be one of the most important
steps in implementing a NetWare network.
Other factors that can affect a tree design include workflow, resource allocation,
network traffic, fault tolerance, and the flow of information through an
organization.

at
TASK 2C-4:

y
op
eM
Discussing tree-design considerations
1. What are some other ways the NDS tree in Figure 2-5 might have been

tC
structured? No
2. What is the main function of the NDS tree?
c

3. What two things should you ensure that a tree design facilitates?
en
Do

4. How can NDS trees be organized?

se
r

5. List some other things that might affect a tree design.

fe
ea
Pl
Re

Objects
In NDS, an object represents a network resource and provides a way to identify
and manage the network resource it represents. You use objects to organize, find,
access, and manage network resources, such as users, servers, and printers.

40
Each NDS object has a name, a list of properties (a collection of information that
describes the object), and specific values for each of its properties.

ial
er
Figure 2-6: An object consists of properties and their values.
In Figure 2-6, the User object BBerkley represents that user on the network and
holds information about him. By configuring the User object’s properties (such as

at
whether it requires a password), a system administrator can control how the user
accesses the network. The object can also store reference information, such as the
user’s phone number.

y
op
eM
Properties and Values
Properties are the categories that can store information for an object. Values are

tC
the information stored in the property fields. NDS stores information about net-
work resources much like information is stored in a database.
No
c
en
Do
se
r
fe
ea
Pl
Re

Figure 2-7: Databases.

Lesson 2: Servers and Clients 41

 Some properties store information that is essential to the network and to the sys-
tem administrator. For instance, a user’s login name is critical in allowing that
user network access. Other properties might contain non-essential information,
NDS defines the properties such as a user’s title, telephone number, or fax number.
for each object. Developers
can use third-party APIs to Some properties have mandatory fields for which you must enter a value. For
develop additional properties example, when you create a User object, Last Name is the only required property.

ial
for an object.
Other properties can hold multiple values. For example, the Telephone Number
property can contain several values (that is, several phone numbers where the
user can be reached).

er
at
y
op
eM
tC
Figure 2-8: An NDS tree showing the [Root] object, Container objects, and Leaf objects.
No
c

Object Types
As its name implies, the NDS tree is a hierarchical tree structure in which NDS
en

objects are stored. NDS objects are divided into three classes: the [Root] object,
Do

Container objects, and Leaf objects.

The [Root] Object

se

During the installation of the first NetWare 4.x or 5.x server on the network, the
r

installation program creates the [Root] object. The [Root] object can contain only
the Novell Directory and Country, Organization, and Alias objects. You must use
fe

square brackets when referring to the [Root] object.

ea

Container Objects
Pl
Re

Container objects enable you to organize the Novell directory logically, much as
folders and drawers organize the documents in a filing cabinet. They can repre-
sent countries, companies, divisional units, departments, responsibility centers,
workgroups, or shared resources.

42
Container objects can hold other Container objects or Leaf objects. Each class of
Container object differs from the others by a set of hierarchical rules that define
what it can contain and where it can exist. The classes also differ in what infor-
mation their properties store.
The following table describes Container objects, including where they can exist
and what objects they can hold.

ial
Icon Name Description Can exist in Can hold
Country Represents the country in [Root] Organization
which the network is located. objects.

Organization Organizes other objects in the [Root] object or Organizational

er
NDS tree. For example, it might Country objects. Unit objects and
represent companies or all Leaf objects.
departments.
Organizational Organizes other objects in the Organization or Organizational
Unit NDS tree. For example, it might Organizational Unit objects and

at
represent departments, Unit objects. all Leaf objects.
business units, project team, or
divisions.

y
op
eM
tC
No
c
en
Do

Figure 2-9: Container objects in an NDS tree.

se
r

See if you can figure out the

container types in this figure.
fe
ea
Pl
Re

Lesson 2: Servers and Clients 43

 Leaf Objects
Leaf objects represent the network’s resources. They cannot contain other objects.
The following table describes the Leaf objects created by default during server
installation:
You can use Novell online
documentation to search for
Icon Leaf object Description

ial
the phrase Leaf objects. The
results of this search will list NetWare Server A server running any version of NetWare starting at 4.x or
Object Classes And higher.
Properties and this section
includes Leaf Object Classes. User Represents a person who can use your network.
This link displays a list of
Leaf objects and the icons Volume Represents a physical volume.

er
associated with them.

Default NDS Objects

at
The following NDS objects are automatically created during the installation of
NetWare 4.x and 5.x:

y
• A NetWare Server object for the server on which you installed NetWare.

op
•
eM
The SYS Volume object, which contains the NetWare operating-system files.
• Any other Volume objects that you created during installation.
• The User object ADMIN , which initially has complete access to all objects

tC
in the tree. An administrator uses this account to perform the initial network
setup.
No
NDS Utilities
c

NetWare provides several utilities for working with the NDS tree. They include:
en

Bindery-based NetWare
Do

operating systems use the

SYSCON utility to manage
users and groups,
PCONSOLE to manage print
services, and FILER to
manage the file system.
se
r
fe
ea
Pl
Re

44
• NetWare Administrator (NWADMN32.EXE), which provides a Windows-
based graphical environment in which you can create NDS objects and
manage the Novell directory. You can also use it to perform printing and
file-management functions. This file is located in the SYS:PUBLIC\WIN32
directory.

ial
er
at
y
op
eM
Figure 2-10: Viewing the NDS tree by using NetWare Administrator.

tC
• NDS Manager, a utility for performing advanced database-management tasks,
such as partitioning and replicating the Novell directory.
• ConsoleOne , a Java application for managing network resources is included
with NetWare 5. ConsoleOne is platform-independent, which means that you
No
can run it from any client or from the server itself.
c

• In NetWare 3.x you use the SYSCON utility used instead of NetWare Admin-
istrator or ConsoleOne to administer the bindery objects (users and groups).
en
Do

Partitions and Replicas

Part of administering the NDS database is making the database secure. This is
done by dividing the NDS database into more manageable pieces, called parti-
tions, and providing copies of those pieces, called replicas, on servers throughout
se
r

the network. Proper placement of these partitions and replicas will help optimize
NDS management, workgroup management, and user access; provide a backup of
fe

NDS, and reduce unnecessary traffic; and eliminate any single points of failure.
ea

Proper partitioning and replication is part of completing your NDS design

structure.
Pl
Re

Partitions
Partitions are logical sections of the NDS database. The NDS database can
become very large. By sectioning the NDS database and placing these sections on
different servers, you can manage the database more efficiently.

Lesson 2: Servers and Clients 45

 Replicas
A replica is a copy of all the data (objects and all their information and values)
stored on a server within the boundary of a partition. Replicas are hidden files
stored on the SYS volume on servers throughout the network and are managed
by administrative users. You should have at least three copies of each partition.
(If you have only one or two servers, you cannot follow this guideline.) Each

ial
partition should have no more than five replicas. Over-replication takes up too
much bandwidth, due to replica synchronization.
Replicas serve two primary purposes:
• Provide fault tolerance; and
• Enhance network performance.

er
Replica Types
There are four types of replicas:

at
• Master replica . The original and only complete copy of a partition. Users
can authenticate using a master replica.

y
• Read/Write replicas. A replica designed for users to view and search for
object information. Users can authenticate using a read/write replica.

op
•
eM
Read-Only replicas. A replica designed for users to view and search for
object information.
• Subordinate references. A pointer created by NDS to point to other replica

tC
types where parent partitions exist without their child partitions.

TASK 2C-5:
No
c

Examining objects in the NDS tree

en

1. Log in to the 411SRVR as Admin. This is the classroom NetWare 4.11

Do

server. Admin is the default administrative user on a NetWare 4.11 network.

2. Start NetWare Administrator by choosing Start→Run, then browsing to

the 411SRVR, then to the PUBLIC\WIN95 folder, and running
NWADMN95.EXE.
se
r

3. Identify the container objects in the browser window. Double-click on

the container objects to expand them and display the Leaf objects below
fe

the containers.
ea

4. Identify the Leaf objects in the browser window. Double-click on a Leaf

object. The properties of the object are displayed in a separate window.
Click Cancel to return to the Browser window.
Pl
Re

5. Explore the properties and values for the Admin object, making sure not
to make any changes to the object.

46
6. Let’s create a user. Select the NetPlus container. Choose Object→Create.
In the Class Of New Object list box, select User, and then click OK. In the
Login name text box, enter your name; then in Last Name, enter your last
name. Click Create. Verify that the user is in the NetPlus container.

7. Close NetWare Administrator and log out.

ial
Summary
In this lesson, you’ve examined the three most popular network operating

er
systems, identified features of network clients, and examined the directory
services used by NT and NetWare networks.

at
LESSON 2 REVIEW

y
2A What are three popular network operating systems?

op
eM
2B What role does the client have in a client-server network?

2C What goals do directory services try to attain?

tC
No
c

What is the Windows NT 4 directory service called and how is it imple-

mented?
en
Do

What is the Novell NetWare directory service called and how is it imple-
mented?
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 47

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Introduction to the OSI LESSON
Model
3

ial
Data Files:
Network+ Certification
Overview CD-ROM

er
In preparation for taking the Network+ exam, you need to have an under- Lesson Time:
standing of how various devices interrelate. The OSI model breaks these 1 hour
communications into layers. This lesson explains those layers and describes
their functionality.

at
Objectives

y
To grasp the details of the OSI model and its relation to most networking compo-

op
nents, you will:
eM
3A Identify the OSI layers.

tC
You’ll examine the importance of the individual layers of the OSI model.
The Application, Presentation, Session, Transport, Network, Data Link,
and Physical layers are the seven layers in the model.

3B Describe relationships between families of protocols.

No
c

You’ll identify the types of protocols that are found in Network and
Transport layers, and in the top three layers (Application, Presentation,
and Session).
en
Do
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 49

 TOPIC 3A
Introduction to the OSI Model
The OSI (Open Systems Interconnect) reference model was developed in the

ial
early 1970s by the International Standards Organization (ISO). The OSI model
was designed to provide a structured and consistent approach for describing,
understanding, and implementing networks. Seven individual layers make up the
structure of the OSI model. Those seven layers in order from highest to lowest
are Application, Presentation, Session, Transport, Network, Data Link, and
Physical. Before the implementation of the globally recognized OSI model, the

er
only solution for an enterprise wide network was to use a single vendor. The OSI
model provides us with a set of general design guidelines for data-
communications systems and also gives us a standard way to describe how
various portions (layers) of data-communication systems interact. Vendors now
use the OSI specifications as a guideline to their network products.

at
y
op
eM
tC
No
c
en

Figure 3-1: Data flow on a simple network architecture.

Do

The various layers represent the data flow pattern that will take place on even the
most basic computer network. Communication takes place between the various
layers in both directions as data moves up or down through the architecture (see
Figure 3-1). Data starts at the Application layer and is sent down through the lay-
ers to the Physical layer. The receiving computer picks up the data at the Physical
se
r

layer and sends it up to the Application layer; this process is repeated over and
over.
fe
ea
Pl
Re

50
Layered Communication
The various layers in a network architecture represent the different stages and
types of processing that occur when information moves through a computer
network. Consider the analogy shown in the following figure, in which Albert
sends a letter to a colleague in Germany. Certain activities occur in each layer to
ensure that Albert’s message is delivered accurately, correctly, and quickly. Com-

ial
munication occurs between the layers and the message moves down or up
through the architecture.
Be sure to notice that when
the sender addressed the
letter to Germany he did
request a return receipt. After
the letter is delivered, a

er
receipt will be returned to the
sender to acknowledge
successful delivery. This can
also occur in network
environments.

at
y
op
eM
tC
No
c
en
Do

Figure 3-2: In network architectures, messages pass through several layers, as in this
analogy.

The Seven Layers of the OSI Model

se
r

Each of the seven layers in the OSI model have a specific set of responsibilities
fe

that allows for reliable network communication. These layers and their general
ea

purpose are described in the following figure.

Pl
Re

Lesson 3: Introduction to the OSI Model 51

 ial
er
at
y
op
eM
Figure 3-3: Layers of the OSI model.
tC
The above layers are numbered from 1 for the Physical layer to 7 for the Appli-
No
cation layer. The responsibilities of each of the layers are briefly summarized
c

below.
Remembering the seven layers can be the toughest thing to do in order to learn
the model. Perhaps the best method of learning these layers is to come up with a
en

memorable mnemonic device. Probably the two best examples of mnemonic

Do

A mnemonic device for devices are All Pizzas Seem To Need Double Pepperoni , and All People Seem To
remembering the layers in Need Data Processing . It may be more beneficial to come up with your own
their proper order is All
mnemonic device so you can remember it come test time. Make it something
Pizzas Seem To Need Double
Pepperoni. funny or catchy and you’re bound to have a great memorization tool.
se
r

Responsibilities of Network Layers

fe
ea

As you learn the responsibilities of each layer, the most important concept to
keep in mind is that each of the layers performs distinct functions, working with
the layers above and below it to enable network communications. Any one layer
on its own is largely useless. And, no layer infringes on the responsibilities of
Pl
Re

any other layer—they don’t do each other’s jobs.

It’s also important to remember that the OSI model layers are only architectural
models and don’t actually perform tasks. Rather, protocol schemes perform the
tasks that correspond to each layer.

52
Same-stack Interaction
The hierarchical layering of protocols on a computer that forms the OSI model is
known as a stack. Typically, each computer has one. Protocols for each layer in a
computer’s stack communicate with their counterparts in another computer’s
stack. To communicate, however, messages must flow down through the lower
layers of the sending stack. A given layer in a stack sends commands to layers

ial
below it and services commands from layers above it. The layers use
preconfigured addresses, known as service access points, to communicate infor-
mation to each other.

Inter-stack Interaction

er
When layers in two different stacks communicate, a layer in one stack communi-
cates with its counterpart in the other stack. It does so by including its own
information with the data information it sends. This information is called a
header, which includes information and any requests by the layer. The layer’s

at
counterpart in the other stack receives, reads, and then processes the header, as
shown in Figure 3-4.

y
op
eM
tC
No
c
en
Do
se
r

Figure 3-4: Sending layers communicate with their counterparts on the receiving stack by
fe

using headers.
ea

When a layer sends its header and the data information to the layer below it, the
lower layer treats all the information as one data packet and attaches its own
header to it. When the data packet reaches its destination in the other stack, the
Pl

original layer’s counterpart recognizes which header it should acknowledge

Re

because lower-layer headers have been stripped off the packet by those layers.
Unlike the downward flow of the data packet in the sending stack, the flow of the
packet in the receiving stack is upward. The data flows up the layers with each
header being stripped off in a first-in, last-out fashion.
In summary:

Lesson 3: Introduction to the OSI Model 53

 • As information moves down the layers, each layer might add its own header
to tell the corresponding layer at the destination how to deal with the data.
By the time the information reaches the Physical layer, five headers could
have been added to it (one for each layer except for Application and
Physical).
• As information moves up the layers at the destination, each layer removes

ial
the header that was added by its peer (the same layer) on the transmitting
end and processes the information contained within that header. When the
information reaches the Application layer, it should be the same as the origi-
nal message. The primary purpose of headers is to enable transmitting layers
to communicate with their peer layers on the receiving end.

er
Service Data Units
The data transferred between layers is known as a service data unit, preceded by
a given layer name. For example, data information given to layer 3 from layer 4

at
is called a Network layer service data unit.
service data unit:
The data transferred between Service data units are made up of data bits, which are referred to by different
names, depending on which layer you are talking about. The following list con-

y
layers of the OSI model.
tains widely used names and the layer to which they pertain. These names,

op
however, are not set standards and are often misused.
eM
• Bits: Physical layer
• Frames: Data Link layer
•

tC
Datagrams and segments: Transport layer
• Messages: Application layer

Note: Packet, among other terms, is also used at many model layers.
No
c

The Application Layer

en

The Application layer includes network software that directly serves the user, pro-
Do

viding such things as the user interface and application features. This layer is
often times mistaken for an actual software package, such as a spreadsheet
program. Do not do this. The Application layer is usually made available by

Application layer:
se
r

Level 7, the highest layer in

the OSI model. It provides a
user interface.
fe
ea
Pl
Re

54
using an Application Programmer Interface (API), or hooks, which are made
available by the networking vendor. This layer doesn’t include application soft-
ware such as Microsoft Word or Excel. However, it does enable commercial
software to use network services such as printing and message handling. Application
Programming Interface
(API):
Application Programming

ial
Interface. An extension to a
programming language that
defines an interface to an
operating system. If an
application is written for a
particular API, it need only
be recompiled to run on any

er
other operating system that
provides that API.

at
y
op
eM
tC
Figure 3-5: The Application layer.

The Presentation Layer

No
c
Software in the Presentation layer attempts to resolve differences in the way data
is represented (character encoding, bit order, and byte order issues) among net-
work computers, for not all computers represent data in the same way. This layer
translates data to ensure that it is presented properly for the end user (in the
en

Application layer). The Presentation layer also handles related issues such as data
Do

encryption and compression, and how data is structured, as in a database. Presentation layer:
Layer 6 in the OSI model. It
provides code and syntax.
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 55

 ial
er
at
y
op
eM
Figure 3-6: The Presentation layer.

The Session Layer

tC
Session layer software establishes and maintains the connection between applica-
tions that communicate with each other through the network. To do this, the
Session layer must employ services provided by the Transport layer. The Session
No
layer comes into play primarily at the beginning and end of a transmission. At the
c

beginning of the transmission, it makes known its intent to transmit. At the end
Session layer: of the transmission, the Session layer determines if the transmission was
Layer 5 in the OSI model. It successful. This layer also manages errors that occur in the upper layers, such as
en

provides a logical a shortage of memory or disk space necessary to complete an operation, or

connection.
Do

printer errors. Network computers can interact on a hierarchical, client/server, or

peer-to-peer basis. Typically, the type of relationship that exists between comput-
ers is established in the Session layer.
se
r
fe
ea
Pl
Re

56
 ial
er
at
y
op
Figure 3-7: The Session layer.
eM
The Transport Layer
The Transport layer provides the upper layers (Application, Presentation, and
tC
Session) with a communication channel to the network. To do this, the Transport
layer uses the services provided to it by the lower layers (Network, Data Link,
No
and Physical). The Transport layer is aware of the packet size that lower layers
c

can handle. When sending a message, the Transport layer determines whether the
message is too big or too small to transmit. If it is too big, the message is broken Transport layer:
down into smaller pieces. If the message is too small, the Transport layer looks to Layer 4 in the OSI model.
en

see if there are other small pieces that are going to the same place, and combines This layer is the heart of the
Do

them when sending. When receiving a message, the Transport layer collects and model; it provides reliability.
reassembles any packets, organizing the segments for delivery and ensuring the
reliability of data delivery by detecting and attempting to correct problems that
occurred (such as missing or improperly sequenced packets of data). If the Trans-
port layer cannot correct data-transmission problems, then it sends a message to
se

the upper layers that a problem has occurred.

r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 57

 ial
er
at
y
op
eM
Figure 3-8: The Transport layer.

The Network Layer

tC
Networks, especially large internetworks, might provide more than one path for
moving information from one point in the network to another. The Network lay-
er’s main purpose is to decide which physical path the information should follow
No
from its source to its destination. The path might span the distance between two
c

computers in the same office, or it might span the globe.

Network layer:
Layer 3 in the OSI model.
en

This layer deals with routing

Do

and addressing.
se
r
fe
ea
Pl
Re

58
 ial
er
at
y
Figure 3-9: The Network layer.

op
eM
The Data Link Layer

tC
Most networks require that only one device transmit on the communication chan-
nel at a time. The Data Link layer provides a system through which network
devices can share the communication channel. This function is called media-
access control (MAC) . The Data Link layer also uniquely identifies each
No
computer on the network, usually by a pre-encoded address that is part of the
c

unit’s network interface card. This pre-encoded address is typically called a MAC Data Link layer:
address and is said to be ″burned in″ by the manufacturer, as they are the ones Layer 2 in the OSI model. It
who do the pre-encoding. shares the media and
en

controls the flow.

The Data Link layer prepares data for transmission over the communication
Do

channel. On the transmitting end, the Data Link layer arranges signals into logical
units of information called frames , which it passes to the Physical layer for
transmission through the network. To ensure that frames are an appropriate length
for the Physical layer, the Data Link layer can fragment data that comes from the
layers above it. On the receiving end, the Data Link layer reassembles the frag-
se
r

mented data and attempts to detect and correct transmission errors that occurred
in the Physical layer. The Data Link layer also establishes and maintains the con-
fe

nection between one device to another while transmission is taking place. This is
ea

called logical link control (LLC) .

To represent the two major functions of the Data Link layer, it is most often con-
sidered to be divided into two sub-layers: the MAC layer and the LLC layer.
Pl
Re

Lesson 3: Introduction to the OSI Model 59

 ial
er
at
y
op
eM
Figure 3-10: The Data Link layer.

The Physical Layer

tC
The Physical layer further addresses the hardware used to transmit data over the
network. This includes the voltage used, the timing of data transmission signals,
No
the requirements for establishing the initial communication connection, acceptable
c

connectors, and interfaces to data-transmission media. The Physical layer provides

the electro-mechanical interface through which data moves among devices on the
Physical layer: network. It specifies the type of data-transmission medium used to link network
en

Layer 1 in the OSI model. It devices. It also specifies how the medium interfaces with network devices, and
Do

is the physical and electrical how data is encoded into electronic signals for transmission through the medium.
interface. Finally, the Physical layer specifies what constitutes an acceptable signal.
se
r
fe
ea
Pl
Re

60
 ial
er
at
y
Figure 3-11: The Physical layer.

op
eM
Making Connections
The Physical and Data Link layers provide local connections, which connect a

tC
node to other devices on the same network segment.
The Network and Transport layers enhance the services provided by the two
lower layers to provide remote connections or internetwork connections, which
link nodes. This type of connection is sometimes called an end-to-end connection.
No
c

The Session layer enhances the services provided by the lower four layers to pro-
vide logical connections, which establish a link between communicating processes
on separate nodes, as shown in Figure 3-12.
en
Do
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 61

 ial
er
at
y
op
eM
Figure 3-12: Making connections.

tC
A Network in Action
As mentioned previously, the protocols in the OSI model layers communicate
with each other to complete tasks. It is similar to sending a letter to a friend. The
No
communications involved in that transaction extend well beyond you and your
c

friend. For example, when you place the stamped and addressed letter in your
mailbox, you communicate to the postal carrier that you want the letter taken to
the post office for sorting and eventual delivery to your friend. The postal carri-
er’s placement of the letter in a bin at the post office communicates to the postal
en

clerk that the letter needs to be sorted. Several other transactions need to occur
Do

before your letter reaches its destination.

Figure 3-13 depicts a simple exchange in which a network node requests a file
from a file server. Network components are shown in relation to the OSI refer-
ence model. As shown here, it is not unusual for some components to span more
se

than one layer of the OSI model to accomplish the communication task at hand.
r

A. An application running on the workstation requests a file that is stored on

fe

the network.
ea

B. Redirection software on the workstation determines that the request is for

network services. It directs the request to the communication protocol for
transmission on the network.
Pl
Re

C. The communication protocol determines the best route through the network
and passes the request to the LAN driver, along with instructions that the
message should be delivered to the file server.
D. The LAN driver copies the request into frames (bundles that can be moved
across the network) and sends the frames to the network adapter for
transmission.

62
E. The workstation’s network interface board transmits the frames across the
network.
F. The file server’s network interface board receives the frames and passes
them to its LAN driver.
G. The file server’s LAN driver takes the message out of the frames and sends
a confirmation to the workstation that it has received the frames.

ial
H. The communication protocol verifies that the message was received intact,
and it passes the message to the network operating system.
I. The network operating system receives the request and acts on it. The net-
work operating system will use the layers of the network to transfer the
requested file from a network hard disk to the workstation.

er
at
y
op
eM
tC
No
c
en

Figure 3-13: Components in a simple exchange between a workstation and a file server,
Do

shown in relationship to the layers of the OSI model.

se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 63

 TASK 3A-1:
Viewing a model of a network in action
Objective: To view the presentation (on the CD included with your
manual) that shows the layers of the OSI model and how the

ial
OSI model layers interact.
Setup: Access to the Network+ Certification CD-ROM should have
been configured on your workstation prior to class.

1. Following your instructor’s directions, access the Network+ Certification

er
CD-ROM. By default , choose Start→Programs→Network+ Interactive
Learning→Network+ Interactive Learning CD.

2. From the Main Menu, click on the Digital Video link.

3. Click the Network Cables And Devices link, and then the OSI Model

at
link.

4. Click on the Overview link to start the presentation.

y
5. View and listen to the digital video presentation.

op
eM
6. Return to the Main Menu and Exit the Network+ Certification
CD-ROM, back to your desktop.

The Benefits of the OSI Model

tC
By separating the tasks necessary for network access into a number of distinct
No
layers, the OSI Model enables the upper layers to work independently of the
c

lower layers. For users, this translates (at least theoretically) to network
transparency. Due to layering, your application should work equally well over an
Ethernet network as over a Token Ring or ARCNet network. Other than speed
en

differences, the networks should work identically because the OSI Model enables
Do

the components (the implementations of a layer or layers) to work independently.

Again, in theory, on a workstation to switch from an Ethernet to a Token Ring
network, you should have to switch only the network interface card (which takes
care of the Physical and some of the Data Link functions) and some of the driv-
ers (to take care of the remainder of the Data Link layer). The rest of the layers,
se

and the drivers that implement them, should be able to remain in place without
r

modification.
fe

The ISO’s publications describe the specifics of each layer in great detail. Even
ea

so, variations exist between network implementations that purportedly follow the
OSI Model. Only your testing can ensure that OSI-compliant systems are truly
compatible for your environment. However, on their own, the ISO’s detailed
specifications go a long way toward ensuring compatibility.
Pl
Re

64
TASK 3A-2:
Identifying layers of the OSI model
1. List the seven layers of the OSI model and the function of each layer.

ial
er
at
2. Create a mnemonic device to help you remember the layers (in order) of

y
the OSI model.

op
eM
tC
APPLY YOUR KNOWLEDGE 3-1 Suggested time:
10 minutes
Developing an analogy for the OSI model
No
Using the analogy in Figure 3-2 as a model, create your own analogy for the OSI
c

model and identify the functions of each layer. The layers in your analogy should
correspond to OSI’s layers. Don’t use data communications for your analogy.
Be sure to include what would happen at each layer of the OSI model:
en
Do

Application layer
Presentation layer
Session layer
Transport layer
se
r

Network layer
Data link layer
fe

Physical layer
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 65

 TOPIC 3B
Families of Protocols
Network communication is provided through protocols —rules by which network

ial
operations are conducted. Network protocols are like species of animals: there are
many of them, and they vary endlessly. Like animal species, protocol ″species″
can be classified into general families that share common features. Some proto-
cols are hybrids; they perform more than one function. Others perform only one
particular function. Computer technology professionals use the OSI model to clas-
sify protocols by function. The following sections introduce families of network

er
Make sure that you protocols and their respective OSI model layers.
understand that OSI model
layers don’t perform network
functions; rather, the
protocols associated with
each layer perform them.
Network and Transport layers

at
The Network and Transport layers provide the following protocol families:
• Reliability protocols provide a way to ensure that reliable data transfer
occurs. For example, a header or trailer might contain a checksum or a

y
request that received data be acknowledged by sending an acknowledgment
message back to the sender.

op
•
eM
Connection protocols are used for establishing and maintaining a
connectionless or connection-oriented service for upper layers. In a
connection-oriented service, the sending and receiving stations maintain con-

tC
stant communication to mediate the transfer of data. Sequencing, flow
control, and reliability are monitored by both ends. This is like a telephone
call in which two parties interact constantly. Connectionless service is more
like the postal service. The message is transferred with little, if any, commu-
No
nication between the two parties, other than the message itself. The message
c

is packaged, delivered, and sent.

• Routing protocols provide a way to ensure that data is transferred to the cor-
rect destination. In an unswitched network, routing is virtually unnecessary,
en

because nodes are directly connected. In a switched network, however, the

Do

routing function determines which path a data unit will take to reach its
destination. This function is particularly important and complex in a packet-
switched network, because there can be many possible paths to a destination
and many intermediary nodes (called routers ) along the way. Routing proto-
cols determine the strategies used to transmit data through the network.
se
r
fe

Application, Presentation, and Session Layers

ea

The Application, Presentation, and Session layers provide the following protocol
families:
• Terminal-emulation protocols enable computers to behave like a standard
Pl
Re

terminal so that they can access a host. This typically involves translation of
keyboard codes and video-display codes.
• File-access protocols enable nodes to use network files. For example, differ-
ent network clients might use different file- and path-naming conventions.
File-access protocols provide a common means to access network files.

66
• File-transfer protocols copy files between network storage and other storage,
such as a computer’s local disk drives.
• Email protocols provide for electronic-mail delivery and handling of related
messages.
• Remote-action protocols determine whether processes should be performed
by a client node or by a server. These protocols are required for setting up a

ial
client-server relationship.
• Network-management protocols provide tools for setting up and maintaining
the network. As networks interconnect into internetworks and become more
complex, more sophisticated network-management tools are necessary.
• Task-to-task protocols enable software processes to communicate over the

er
network.
• Multiple-session protocols enable multiple network links to be established.
• Codeset-and-data-structure protocols define how data is represented. These
protocols translate data for nodes that use different coding schemes.

at
TASK 3B-1:

y
Identifying families of protocols

op
eM
1. Identify the family of protocols that determines whether a client or
server completes a process.

2. Identify the family of protocols that ensures that data is transferred

without errors.
tC
No
c

3. Identify the family of protocols that enables a computer to access a host

node by behaving like a terminal.
en
Do

4. Identify the family of protocols that provides tools for setting up and
maintaining the network.
se
r

5. Define the function of the routing family of protocols.

fe
ea

6. Define the function of the file access family of protocols.

Pl
Re

Lesson 3: Introduction to the OSI Model 67

 7. Define the function of the codeset and data structure family of protocols.

8. Define the function of the connection family of protocols.

ial
Summary

er
In this lesson, you’ve identified the layers of the OSI reference model and
learned the functions of each of those layers. You also learned about the
kinds of protocols that work at each of the OSI layers.

at
LESSON 3 REVIEW

y
op
eM
3A Describe the purpose of network architecture and the OSI model.

tC
List the OSI layers and describe how they communicate.
No
c

3B Describe what is meant by the families of protocols.

en
Do
se
r
fe
ea
Pl
Re

68
The Physical Media LESSON
4

ial
Overview Data Files:
This lesson will concentrate on the Physical layer of the OSI model. The Network+ Certification
Physical layer describes the physical and electrical characteristics of the net- CD-ROM

er
work medium. It deals with actually putting bits onto the communication
medium. This lesson will give you detailed descriptions of various compo- Lesson Time:
nents of a computer network that correspond to the Physical layer of the 1 hour
OSI model.

at
Objectives

y
To understand the components of a computer network that correspond to the

op
Physical layer of the OSI Model, you will:
eM
4A Examine bounded data-transmission media and connectors.
You’ll examine network cabling such as twisted-pair, coax, and fiber optic

tC
cables. You’ll also examine the advantages and disadvantages of each and
determine when you should use them.

4B Examine unbounded data-transmission media and connectors.

No
You’ll examine unbounded media including radio, microwave, and laser
c

transmissions. You’ll also compare these to some of the bounded trans-

mission media.
en

4C Distinguish between analog and digital signaling.

Do

This topic covers use of switches to represent digital values (binary val-
ues) and differentiate between digital and analog devices.

4D Compare baseband and broadband transmissions.

You’ll examine these two signaling techniques and identify where each is
se
r

used.
fe
ea
Pl
Re

Lesson 4: The Physical Media 69

 TOPIC 4A
Data-transmission Media
For computers to communicate, they need to be connected. The data-transmission

ial
media is the channel or conduit through which data is transmitted. Computers can
be connected by many different kinds of data-transmission media, which can gen-
erally be divided into two categories: bounded and unbounded. You can think of
bounded media as wires or cables and unbounded media as the airwaves. This
topic introduces the most common bounded media, including twisted-pair cable,
coaxial cable, and fiber optic cable; and unbounded media, including radio, laser,

er
bounded media: and infrared.
A physical product, such as
cable, through which data is You should be aware of the characteristics of each medium when deciding on the
transmitted. best one for your network needs. There are physical limits with each type of
medium. For example, attenuation, or weakening of a signal over distance, is a
factor in many types of transmission media. The amount of data that can be car-

at
ried on any particular transmission medium is referred to as bandwidth.
attenuation:
Degradation of signal. Bandwidth is measured in Hertz (cycles per second) or bits per second (bps).

y
Although there are other considerations in a network’s performance, the general
rule is that the greater the bandwidth, the greater the amount of data that can be

op
eM
transmitted. How fast the data moves within the medium is referred to as the data
rate , and it is also measured in bps. Don’t confuse bandwidth with data rate.
bandwidth:
The difference between the One way to think of it is that bandwidth can be compared to the number of lanes
highest and lowest on a highway and the data rate is the speed for the highway. Bandwidth and data

tC
frequencies used for a rate combined equal throughput. A data-transmission medium with a high band-
communication channel.
width and a low data rate could have the same throughput as a data-transmission
Generally, more bandwidth
means greater transmission medium with a low bandwidth and a high data rate.
capacity.
No
c

data rate:
How fast the data moves
en

within the medium.

Do

throughput:
A measure of productivity on
se

a device or network.
r
fe
ea
Pl
Re

Figure 4-1: Bandwidth versus data rate.

70
Bounded Media
Bounded media transmit signals by sending electricity or light over a cable.

Twisted-pair Cable

ial
In its simplest form, twisted-pair cable (TP) contains a pair of insulated copper
wires that are twisted around each other and encased in a protective sheath. The
following figure illustrates the two major types of twisted-pair cabling: unshielded
twisted-pair (UTP) and shielded twisted-pair (STP). In STP, the inner wires are
encased in a sheath of foil or braided wire mesh.

er
at
y
op
eM
twisted pair:
Cable consisting of at least
two insulated wires that are

tC
intertwined to reduce
electromagnetic interference.
No
Figure 4-2: UTP and STP cables.
c

unshielded twisted pair:

Twisted pair wire that isn’t
protected from EMI through
the use of shielding.
en
Do

shielded twisted pair:

Twisted pair wire that’s
surrounded by a foil or
braided metal sheath for the
se
r

purpose of reducing
susceptibility to
fe

electromagnetic interference.
ea
Pl
Re

Lesson 4: The Physical Media 71

 In both UTP and STP, the wires are twisted around one another to reduce electro-
magnetic interference (EMI). EMI can occur as noise (interference from outside
sources such as power sources) or as crosstalk (signal overflow from an adjacent
wire), and can severely hinder data communications. The shielding of STP is
EMI: designed to further reduce its susceptibility to interference.
Electromagnetic interference.
Noise or crosstalk that can The number of pairs of wires contained in a twisted-pair cable varies. Also, there

ial
severely hinder data are standards governing how many twists are required per foot of cable. Most
communication. twisted-pair cable is relatively inexpensive. Twisted-pair cable has been used for
years in the telephone industry, so the technology and standards for this type of
cable are well established. However, you should be aware that many standard
telephone cables aren’t twisted pair and are inadequate for most network
crosstalk: applications.
Signal overflow from an

er
adjacent wire. The Electronics Industries Association and the Telecommunications Industries
Association’s (EIA/TIA) 568 Commercial Building Wiring Standard defines five
categories of UTP, which are described in the following table.

Category Description

at
Category 1 Traditional UTP telephone cable. Category 1 can transmit voice signals but not
data. Most telephone cable installed prior to 1983 is Category 1.

y
Category 2 UTP cable made up of four twisted-pair wires, certified for transmitting data up to
4 Mbps (megabits per second).

op
Category 3
eM UTP cable made up of four twisted-pair wires, each twisted three times per foot.
Category 3 is certified to transmit data up to 10 Mbps.
Category 4 UTP cable made up of four twisted-pair wires, certified to transmit data up to 16
Mbps.

tC
Category 5 UTP cable made up of four twisted-pair wires, certified to transmit data up to 100
Mbps.
No
c

The IBM Cabling System

In 1984, IBM introduced its own cabling system, complete with standards and
en

specifications. IBM developed custom cable connectors, face plates, distribution

Do

panels, and more. Much of this specification was designed to support IBM’s net-
working technology, Token Ring.

IBM type Standard label Description

AWG:
Type 1 Shielded twisted pair Two pairs of 22 AWG (American Wire Gauge standard*)
se

A measurement of wire
r

diameter, where the diameter wires surrounded by braided wire shield.

of the wire increases as the Type 2 Voice and data cable Two pairs of 22 AWG wires surrounded by braided shield
fe

AWG number gets smaller. used for data surrounded by four additional pairs of 26 AWG
ea

Thus, AWG 22 wire is thicker cables used for voice.

than AWG 26 wire.
Type 3 Voice grade cable Four, solid, unshielded twisted-pair 22 or 24 AWG wires.
Type 4 Not defined
Pl

Type 5 Fiber optic cable Two 62.5/125 micron multimode optical fibers.
Re

Type 6 Data patch cable Two pairs of 26 AWG stranded wires with a dual foil and
braided shield.
Type 7 Not defined
Type 8 Carpet cable Two shielded twisted pairs of 26 AWG cables housed in a
flat jacket (designed to be run under carpets).

72
IBM type Standard label Description
Type 9 Plenum cable Two shielded twisted-pair cables in a fire-safe jacket.

* AWG is a measurement of wire diameter, where the diameter of the wire

increases as the AWG number gets smaller. Thus, AWG 22 wire is thicker than

ial
AWG 26 wire.

Twisted-pair Connectors
A common connector type used with twisted-pair cable is the RJ-45 connector.

er
(Certainly, other types of connectors are used, though less commonly than the
RJ-45.) The RJ-45 connector is similar to a typical telephone plug (RJ-11); how-
ever, the RJ-45 is larger. It contains eight cable connections compared to the
RJ-11, which contains only four.
RJ-45:
A connector used with

at
twisted-pair cable that
contains eight cable
connections.

y
op
eM
tC
Figure 4-3: RJ-45 connectors.
Additionally, large twisted-pair networks can include the following hardware
components:
No
Component Purpose
c

Distribution rack and rack Used to create more room for cables where floor space is at a
shelves premium, and to centralize and organize network connections.
Expandable patch panel Support for multiple ports. Transmission speeds of 100 Mbps.
en
Do

Jack couplers (patch cord) Single or double RJ-45 jacks that snap into patch panels, and
wallplates to connect two ports, such as the computer to the
wallplate. They support data rates to 100 Mbps.
Wall plates Supports two or more jack couplers.
se
r
fe
ea
Pl
Re

Lesson 4: The Physical Media 73

 ial
er
Figure 4-4: Twisted-pair patch panel.
STP is capable of handling more data with greater speed than is UTP. Twisted-
pair cable is susceptible to eavesdropping; however, STP is less susceptible than

at
UTP. The following table lists the advantages and disadvantages of twisted-pair
cabling.

y
Advantages Disadvantages

op
Is inexpensive.
eM
Has well-established products
Is susceptible to EMI and eavesdropping.
Has a low bandwidth.
and technology.
Is easy to install.

Crimping RJ-45 Connections

tC
No
c

To create an RJ-45 modular cable and properly terminate it, you’ll most likely
use the T568A wiring scheme. This scheme was created by the EIA/TIA (Elec-
tronic Industry Association/Telecommunications Industry Association). T568A
en

uses 8 wire twisted-pair cabling with the following wiring scheme:

Do

Pin Wire color Ethernet Token Ring

1 White/Green Transmit+ ----
2 Green Transmit- Ground
3 White/Orange Receive+ Transmit+
se
r

4 Blue --- Receive+

5 White/Blue --- Receive-
fe

6 Orange Receive- Recovery

ea

7 White/Brown --- Ground

8 Brown --- ---
Pl
Re

74
The T568A wiring scheme can be used for both 10BaseT and Token Ring
networks. To make an RJ-45 modular cable:
1. Slit the PVC covering on the cable about 1.5 cm from the end, being careful
not to cut or damage any of the internal wires.
2. Spread out the wires in the order shown in the T568A table, keeping them
parallel and close together.

ial
3. Clip the wires so they are even in length.
4. Slide the wires into the modular plug with the White/Green wire on the left.
Make sure the wires are all the way in under the metal teeth in the plug.
5. Insert the plug into the crimper tool while not getting any slack in the wires.
Close the handles on the crimper hard so that all of the teeth in the plug bite

er
through the wires.
6. Check the plug by looking to see if you see any wires that weren’t bitten.
Gently pull on the cable to see that it’s firmly crimped. You can also use a
cable tester to check continuity.

at
Coaxial Cable

y
Coaxial cable, most often called coax , contains a single-wire inner conductor

op
eM
surrounded by a layer of insulation, a shield of braided wire mesh or foil, and an
outer layer of insulation. The inner conductor carries the data and can be either
solid (usually copper) or stranded. The following figure illustrates typical coax
cable.

tC
coaxial cable:
A data-transmission medium
that contains a single-wire
inner conductor surrounded
by a layer of insulation, a
No
shield of braided wire mesh
c

or foil, and an outer layer of

insulation.
en

Figure 4-5: Coaxial cable.

Do

The purpose of the shield, typically a braided wire mesh, is to act as a ground to
protect data from EMI. If the core and the shield touch, the cable will short, and
noise or stray data signals will flow onto the core, destroying the data signal
being transmitted. Thus, there must be a layer of insulation between the core and
se

the shield. The entire cable is wrapped in another layer of insulating material,
r

usually made of rubber, Teflon, or plastic. Coax gets its name because of the fact
that the conductor and the shield share the same axis, thus are ″co-axial.″
fe
ea

There are two types of coaxial cable: thin (thinnet) and thick (thicknet) as shown
in the following figure.
Pl
Re

Lesson 4: The Physical Media 75

 ial
er
Figure 4-6: Thicknet versus thinnet coaxial cable.
The following table provides a comparison of thinnet and thicknet coaxial cable.

Thinnet coaxial cable Thicknet coaxial cable

at
Can carry a signal up to approximately 185 Can carry a signal up to approximately 500 meters
meters (607 feet). (1,640 feet).
Is about 0.25 inches thick. Is about 0.50 inches thick.

y
Is flexible and easy to work with in connecting Is thicker than thinnet and is used mainly as a
network computers. ″backbone″ to connect thinnet networks together.

op
eM
Coax cable can be further categorized based on a number of specifications:

tC
Coax cable type Impedance Description
The RG stands for Radio
Grade and is an old term RG-58/U 53.5 ohms Solid copper core thinnet cable.
defined by the military RG-58 A/U 50 ohms Stranded copper core thinnet cable.
specifications that originally
No
RG-58 C/U 50 ohms The military version (MILSPEC) of RG-58 A/U.
designated these
c

specifications. RG-8 50 ohms Stranded core thicknet (broadband) cable.

RG-11 75 ohms Dual-shielded, solid core thicknet cable.
RG-59 75 ohms Dual-shielded, solid core cable used for broadband
en

transmissions, such as cable television. Not suitable for

Do

For more information on most networking applications.

these and other cable
RG-6 75 ohms Similar to RG-59, but larger in diameter. Can handle
specifications, visit the
Belden Web site at www. higher bandwidths than RG-59.
belden.com. RG-62 93 ohms ARCnet networking cable, not suitable for Ethernet
networking.
se
r
fe
ea
Pl
Re

76
Coaxial Connectors
A British Naval Connector (BNC)—also called a Bayonet Nut Connector, Bayonet
Neill Concelman, or Bayonet Navy Connector —is usually used for thinnet
coaxial cable. The connector fits into a coupling and is twisted to lock it into
place. A BNC T-connector provides three-way connection: two connections for
the network and one connection for the transceiver device. At the end of a length

ial
of cable, you will often see terminators. A terminator is a resistor attached to the
end of the cable. Its purpose is to prevent signal reflections, effectively making
the cable ″look″ infinitely long to the signals being sent across it. The thinnet BNC:
connector types are shown in the following figure. (British Naval Connector or
Bayonet Nut Connector)
Connector usually used for
thinnet coaxial cable.

er
Provides three-way
connection—two
connections for the network
and one for the transceiver
device.

at
terminator:

y
A resistor attached to the end
of the cable to prevent signal

op
eM reflections, making the cable
appear to be infinitely long
to the signals being sent
across it.

Figure 4-7: Coaxial connection hardware.

tC
No
c

Thicknet coaxial cable uses a vampire tap to connect to a transceiver. Every few
feet of cable is marked to indicate where a new connection can be made. The
tool to make the connection is a metal spike that pierces the cable. A special con-
en

nector is then screwed into the thick coaxial cable.

Do
se
r

Figure 4-8: A vampire tap, ready to be installed on the cable.

fe
ea
Pl
Re

Lesson 4: The Physical Media 77

 A cable connects the transceiver and the Attachment Unit Interface (AUI) port on
the network adapter. An AUI connector is a 15-pin D-shaped connector (a DB-15
connector) that looks like a parallel port connector. Another commonly used name
for an AUI connector is a DIX connector, named for the three companies that
AUI connector: developed it—Digital, Intel, and Xerox.
(Attachment Unit Interface) A
15-pin D-shaped connector Most specifications and many local electrical codes require you to ground one

ial
(a DB-15 connector) used in point on a coax segment. Such a ground connection allows static electricity to
thicknet cable networks. harmlessly drain off rather than build to the point where people or equipment
could be harmed. Make sure you ground your cable at only one point. Multiple
grounds on an Ethernet segment will cause network errors, and could increase the
risk of damage to equipment or injury to people.
DIX connector:
Another name for an AUI The following table lists the advantages and disadvantages of coaxial cable:

er
connector, named for the
three companies that Advantages Disadvantages
developed it: Digital, Intel,
and Xerox. Is light, flexible, and easy to work with. Can be difficult and expensive to install.
Is not as susceptible to EMI and Is susceptible to EMI and eavesdropping, although

at
eavesdropping as twisted-pair cabling. better than twisted-pair cable.
Is inexpensive. Is more expensive than twisted-pair cabling.
Has well-established products and technology.

y
Is more resistant to attenuation (the weakening
of a signal over a distance); is better than

op
eM
twisted-pair cabling for longer distances.

Crimping BNC Connectors on Coaxial Cables

tC
If you are using thinnet, you’ll need to either purchase pre-made cables, or create
your own. The following steps explain how to terminate BNC connections:
No
1. Start with RG58 coaxial cable, a crimp, pin, and BNC connector.
c

2. Slide the crimp over the coax cable, then using wire strippers, strip off a
length of the outer covering that is slightly shorter than the length of the
en

crimp.
Do

3. Pull the braided layer back over the stripped cable, exposing the insulation
around the center conductor. Be sure not to cut off the braiding.
4. Remove a length of insulator the same length as the bottom of the pin to the
indentation on the pin, then insert the center conductor all the way into the
pin.
se
r

5. Using the crimping tool, crimp the pin to the center conductor, making sure
it is held tightly.
fe
ea

6. Push the BNC down over the pin until it grabs the cable.
7. Slide the crimp up and over the braid and base of the BNC.
8. Crimp the crimp to the cable with the crimping tool.
Pl
Re

78
Fiber Optic Cable
Fiber optic cable transmits signals of light though a very fine strand of plastic or
glass (optical fiber), called the core . The core is encased in a tube that is made
of glass, called cladding . Light bounces from side to side as it moves through
the optical fiber. The cladding is designed to reflect light back into the optical
fiber. The cladding is then surrounded by a tough outer sheath, called a jacket ,

ial
which provides a protective covering for the optical fiber. Each optic fiber core
can transmit data in only one direction, so a cable generally consists of two
strands in separate jackets. One strand would send data, the other strand would fiber optic cable:
receive it. A cable that transmits
signals of light through a
very fine strand of plastic or
glass (optical fiber) called

er
the core, which is encased in
a tube made of glass, called
cladding, which in turn, is
encased in a tough outer
sheath called a jacket.

at
Figure 4-9: Fiber optic cable.

y
op
eM
Fiber Optic Connectors
Fiber optic connectors are used for fiber optic cable. SMA connectors have a

tC
threaded nut that is twisted to lock it into place. ST connectors use a bayonet
locking system.
The following table lists the advantages and disadvantages of fiber optic cable:
No
Advantages Disadvantages
c

Is nearly impossible to tap, making it secure. Is expensive.

Doesn’t produce (and isn’t affected by) EMI. Is more complex to install than other types of
cable.
en
Do

Has a high capacity for transmitting data, generally Has less-established products and technology
100 Mbps or faster (lab experiments have when compared to other bounded media.
demonstrated throughputs of 1 Gbps).
Is lightweight. Is brittle and breaks easily.
Has very little signal attenuation, enabling data
se

transmission for miles.

r

Uses less energy to operate than other transmission

media.
fe
ea

Cable Grades
Pl
Re

Cables are also graded for fire safety. Many cables contain polyvinyl chloride
(PVC), a plastic used to make the insulation in the cables. PVC is flexible, mak-
ing cable made with it easy to install. However, when PVC burns, it creates
poisonous gases. National and local fire codes regulate where PVC cables can

Lesson 4: The Physical Media 79

 exist in a building. One place they cannot run is within a plenum. A plenum is an
air-handling space (part of the heating/cooling system) and is often a convenient
place to run cables.
plenum cable:
Cable made of special Because poisonous gases and flames can spread quickly throughout the building
materials in the insulation via a plenum, a special cable must be used. Plenum cable, which gets its name
layers that make it fire from the plenum space, has special materials in the insulation layers that make it

ial
resistant. When it burns, it fire resistant. When plenum cable burns, it produces a minimal amount of smoke
produces a minimal amount and chemical fumes.
of smoke and chemical
fumes.

er
at
y
op
eM
tC
No
c
en

Figure 4-10: Examples of locations for PVC and plenum cables.

Do

Plenum cable was designed to meet standards set by the National Fire Protection
Agency (NFPA) and others for smoke and toxic vapors in air-handling spaces.
Local fire codes have also affected the design of plenum cabling, requiring that
the cables themselves cannot be a conduit for fire. The special materials in the
se

insulation also make plenum cabling more expensive and less flexible than PVC
r

cable.
fe

Cable rating Standard for performance on a flamespread test

ea

FT4 Flame Test 4 rating as described by the Canadian Standards Association (CSA)
standard C22.2.0.3-1992.
FT6 Flame Test 6 rating described by the National Fire Protection Agency (NFPA)
Pl
Re

standard 262-1985.
UL-910 National Electrical Code (NEC) standard based on Underwriters Laboratories
(UL) tests for smoke and flame spread.

80
Cable classification Description
CMS, CL2X Restricted cable; must be enclosed in a conduit; may have only up to
10 feet of cable exposed; must pass UL test 1581 VW-1.
CM, CL2 General-purpose cable; used in areas other than risers (a vertical shaft
used to route cabling between floors) or plenums; must pass UL
vertical tray test 1581.

ial
CMR, CL2R Riser cable; used in vertical shafts; must pass UL test 1666.
CMP, CL2P Plenum cable; used in plenum areas; must pass UL test 910 for
smoke and flame spread.

er
Comparing Media Types
The data-transmission medium (or media) that a network uses determines many
of the characteristics and limitations of the network. The following are some of
the things you should consider when you compare data-transmission media:

at
• Susceptibility to EMI . EMI can come from a variety of sources, including
fluorescent lights, elevators, water coolers, power supplies and transformers,

y
high-voltage equipment (such as welders), and components of other commu-
nications systems (such as radio transmitters). EMI can severely hinder data

op
communications.
eM
• Security . Not only are many types of data-transmission media susceptible to
EMI, many emit EMI as well. This can create a problem in crowded wiring

tC
closets, where cables can interfere with each other or create a security
problem. A nearby receiver could pick up electromagnetic signals emanating
from a cable. With the proper equipment, those signals could be decoded and
interpreted (wiretapped).
•
No
Bandwidth, or the transmission capacity of the medium . Typically, the
c

greater the bandwidth, the greater the amount of data that can be transferred
over a network.
• Cost to purchase and maintain . Although the initial cost might be low for
en

certain media, you should also consider other expenses that you will experi-
Do

ence over the life of the medium, such as maintenance and upgrade costs.
Also, certain products are more durable than others and might hold up to the
ravages of time, use, and the environment better than others.
• Compatibility with other products in your network . If possible, you should
purchase products that adhere to established standards. It is more likely that
se
r

future networking products will be compatible with standard media than

non-standard.
fe

• Ease of troubleshooting . You should consider the significance of your

ea

network. If it fails, what are the costs of extended down time due to diffi-
culty in troubleshooting. What is the level of expertise in your support
personnel? Do they have the knowledge or experience to troubleshoot a
Pl

more complex physical media?

Re

Lesson 4: The Physical Media 81

 TASK 4A-1:
Matching the cable type with its properties
Objective: For each of the following statements, identify which transmis-
sion media is being described.

ial
1. Can’t be tapped and result in stolen data.

2. A flexible cable that is easily installed, but shouldn’t be placed in crawl

spaces.

er
3. Better for very high-speed, high-capacity data transmission because of
the lack of attenuation and the purity of the signal.

at
4. The most commonly-used cable for existing telephone systems; already

y
installed in many office buildings.

op
eM
5. Uses a foil wrap between and around wire pairs; uses internal twisting
of the pairs to give improved protection from EMI.

APPLY YOUR KNOWLEDGE 4-1

tC
No
Suggested time:
c

15 minutes
Viewing videos about cables and connectors
You’ll use the Certification CD included with your manual to view digital video
en

presentations about cables and connectors.

Do

1. Load the Digital Video component of the Network+ Certification CD-ROM.

2. From the Course list, choose Network Cables And Devices.

se

3. Click on Cables and Connectors, and then click Overview to begin the video
r

presentation. When the presentation has finished, you’ll be back at the topic
list, ready to pick out your next video.
fe
ea

4. Click Crimping RJ-45 And BNC Connectors, and then click Overview to
begin the video presentation.

5. Return to the Main Menu, and exit the Network+ Certification CD-ROM.
Pl
Re

82
Building a Network
Let’s now examine some physical networking components. The following
optional activity lets you find the appropriate network components that go
together to form a LAN.

ial
OPTIONAL TASK 4A-2:
Building a network
Objective: To build a network without any computers (just the network-

er
ing components that would be required to create a network).
Setup: You can complete this task if your instructor has provided net-
work cards, cabling, and any other devices which would be
required to cable together two or more computers in a LAN.

at
1. Identify the network card and associated cable type for your network.

2. Connect the network card and cabling, using any other components you

y
might need.

op
3.
eM
Explain to your classmates what kind of network this would be used for
and what components are included in your network.

4. If other topologies are available, return the components you already used,

tC
and then connect another topology’s components together.

5. Again, explain what kind of network you’ve created and what compo-
nents were included.
No
c

6. Return all of the components.

TOPIC 4B
en
Do

Unbounded Media
Unbounded media transmits data without the benefit of a conduit—it might trans-
se
r

mit data through open air, water, or even a vacuum. In addition to various forms
of radio, unbounded media includes links based on laser, infrared, and other light
forms. Unbounded media, such as infrared, can be used to connect computers or
fe

other devices to a network without a cable, creating a wireless component to your

ea

cabled network.
A wireless network enables a business to:
•
Pl

Provide users with temporary connections to an existing LAN.

Re

• Create a backup to an existing LAN.

• Provide a certain degree of portability to its network, allowing users to work
from remote locations.
• Extend its networks beyond the physical limitations of cabling.

Lesson 4: The Physical Media 83

 • Enable employees, such as salespeople or doctors, to move around
frequently.
• Provide network connections in buildings where cabling would be difficult,
such as buildings designated as historical landmarks.
• Provide network connections in busy areas, such as lobbies, where cables
could be unsightly or even dangerous (people could trip over cables, but are

ial
not likely to trip over radio waves).
Other unbounded media, such as microwave, can be used to connect networks
over very long distances. Typically, unbounded media are more prone to eaves-
dropping and electromagnetic and atmospheric interference (such as rain and fog)
than are bounded media.

er
Radio
Although the term radio can refer to a large portion of the electromagnetic spec-

at
trum, we will refer to the range of frequencies that includes FM radio, television
(VHF and UHF), and computer networks. Transmissions in this frequency range
are limited by distance because they don’t bounce off the ionosphere and aren’t

y
as affected by atmospheric conditions (unlike microwave transmissions). For these
reasons, radio is suitable for data communications. However, because transmis-

op
eM
sion frequencies of radio transmissions tend to be lower than those of microwave
transmissions, data transmissions are constrained by physical objects (like build-
ings and trees). Also, radio transmission systems tend to have slower data rates
than some other types of unbounded media.

tC
There are two types of radio transmissions: narrow-band (single-frequency) radio
and spread-spectrum radio. In narrow-band radio, you tune both the transmitter
and the receiver to a certain frequency. Just as you can receive a music radio sta-
tion inside a building, data transmissions using radio frequencies can pass through
No
narrow-band radio: walls. This means a radio data link doesn’t require line of sight focusing. Typical
c

A radio signal where both radio network devices have a broadcast range of about 5,000 square feet.
the transmitter and receiver
are tuned to a certain Because radio systems transmit at a high frequency, transmissions cannot occur
frequency to send and through steel or load-bearing walls. (Just as you can hear more static than music
en

receive data. inside buildings built with walls of this type.)

Do

Spread-spectrum radio works differently than broadcast radio. While there are two
primary types of spread spectrum radio, by far the most common system is
spread-spectrum radio: frequency-hopping spread spectrum . In this type of system, data is transmitted
A radio signal where the over a range of frequencies, divided into channels , or hops . The transmitter and
se

transmitter and receiver tune the receiver tune into a specific channel for a predetermined period of time; they
r

into a specific channel for a then switch to a different channel, and so on, continuing this pattern until the data
predetermined period of
has been transmitted and received. Eavesdropping is discouraged because only the
fe

time. They then switch to a

sender and receiver know the frequencies selected (out of the millions possible),
ea

different channel periodically

until the data has been the frequency switching order, and each channel’s transmission times. In general,
transmitted and received. spread-spectrum radio is a slow transmission method compared to other
unbounded media.
Pl
Re

84
Typically, frequency-hopping spread spectrum can offer 250 Kbps throughput,
though some implementations can reach 4 Mbps or faster. The range of this sys-
tem is typically two miles outdoors and about 800 feet indoors.
The second type of spread spectrum radio, Direct Sequence Spread Spectrum ,
spreads the signal across many frequencies simultaneously. DSSS combines a Additional information on
radio carrier signal with the signal in such a way that the resulting output is a frequency-hopping and

ial
wide-bandwidth, low-power wave. Signals generated with this technique appear DSSS spread-spectrum radio
as noise to eavesdroppers. This system is complex and expensive. It is sometimes is available at http://www.
used by the military because intercepting this type of transmission is very cwt.vt.edu/faq/ss.htm.
difficult. Range and throughput are limited with this type of system.

Note: You must have a license to use some bands of radio as a network medium,

er
and it is susceptible to EMI and eavesdropping. Many products use specific fre-
quencies set aside by the Federal Communications Commission (FCC) for
unlicensed use. These bands are called Industrial, Scientific, and Medical (ISM)
bands. Just because you might not need a license to use these bands does not
mean that you can pick any radio frequency within these designated ranges. The
FCC regulates specific bands and what they can be used for, even if the FCC

at
doesn’t require a license for such use.

Cellular phone service (an example of radio communications), which uses the

y
ISM bands, can provide an efficient portable network connection. Cellular Digital

op
Packet Data (CDPD) technology enables packet (network style) communications
over the cellular telephone voice network.
eM
tC
Terrestrial Microwave
Terrestrial microwave is commonly used for long-distance voice and video trans-
missions, and for short-distance high-speed links between buildings. A common
No
type of microwave antenna is a parabolic dish about 10 feet in diameter. In an
area of rough terrain, microwave technology is often less expensive than cabling.
c
en
Do
se
r
fe
ea
Pl
Re

Figure 4-11: Terrestrial microwave.

Lesson 4: The Physical Media 85

 Microwave is similar to radio, but a major difference is that radio broadcasts tend
to be omnidirectional (broadcast in all directions), while microwave transmissions
are focused in one direction (to the other dish). For this reason, microwave anten-
nas must be aligned precisely for successful transmission. Transmission from far
beyond the horizon cannot be picked up. The atmosphere bends microwave sig-
nals toward the earth somewhat, extending the limit of microwave transmissions
beyond the line of sight, although not enough to follow the curvature of the

ial
earth’s surface perfectly. The following table shows how elevating the antennas
above the ground can extend the distance between points. The measurements in
feet and miles are approximate.

Elevation of antennas Maximum distance between points

er
10 meters (32 feet) 26.07 kilometers (16 miles)
100 meters (328 feet) 82.45 kilometers (51 miles)
500 meters (1,640 feet) 184.35 kilometers (114 miles)

High frequencies generally correspond to a high capacity for carrying data, so

at
very high capacities are available within the high end of the microwave range.
However, higher frequencies are more susceptible to atmospheric interference like
rain. Also, buildings, motor vehicles, mountains, and other solid objects can

y
obstruct microwave transmissions. Again, it is important to note that a license is

op
required for microwave transmissions and the transmissions are susceptible to
eavesdropping.
eM
tC
Satellite Microwave
To enable networks to traverse distances greater than those provided by terrestrial
microwave, communications satellites high above the earth’s surface can relay
signals between ground stations. When a ground station transmits data to the
No
communications satellite (a process called uplinking ), the satellite relays the data
c

transmission to another ground station (a process called downlinking ), and then

repeats the data. The ground station can be mobile. Typical communications satel-
lites are in a geosynchronous orbit—their path follows the rotation of the earth,
en

enabling them to remain in the same relative location above the earth.
Do
se
r
fe
ea
Pl
Re

86
 ial
er
at
y
op
eM
Figure 4-12: Satellite microwave.
The physics involved dictate that communications satellites must be 22,300 miles

tC
above the earth. Microwave signals move data at the speed of light. Even at this
high speed, the great distance (more than 44,600 miles by the time a data trans-
mission returns to earth) involved in a satellite microwave transmission means
that there is a significant delay between the time data is sent and the time they
No
are received. When combined with the time it takes for the satellite to repeat the
c

signal, the delay is typically 240 to 300 milliseconds. Sometimes you can notice
this delay in long-distance telephone conversations that use satellite microwave
transmissions; the echo of your own voice is slightly delayed. For some network-
en

ing applications, this delay can introduce problems. Satellite microwave

Do

technology is expensive, requires a license, is susceptible to eavesdropping, and is

affected by weather conditions.

Laser
se
r

A laser transmitter uses a highly focused beam of light pulses to transmit data.
The receiver uses a photoelectronic (light-sensitive) device to convert the laser
fe

back into the original data. Like optical fiber, laser uses light, which occurs at a
ea

higher frequency than the frequencies used in microwave and various forms of
radio. Because of this, laser is capable of very high data rates. However, laser is
also susceptible to attenuation (loss of power) and interference, which weakens
Pl

the signal.
Re

Lesson 4: The Physical Media 87

 One popular type of laser networking system is point-to-point infrared. Although
the interface devices are expensive, they are resistant to eavesdropping, and
capable of high data transmission rates. However, laser transmissions are affected
by atmospheric conditions, susceptible to high-intensity light, and require line-of-
sight positioning. Laser (not infrared) is commonly used for building-to-building
links and can reach typical LAN speeds or beyond. This is a low-cost way to link
two buildings, as compared to digging a trench and laying cable.

ial
er
at
y
op
eM
Figure 4-13: Infrared light.
Infrared transmitters are fairly common devices used in many homes—typical

tC
remote controllers for televisions, videocassette recorders, and stereos use infrared
light to transmit data. Infrared systems don’t require a focused beam, yet aren’t
omnidirectional either. Infrared signals reflect easily off many hard surfaces typi-
cal of an office or home. For example, the television remote control can be aimed
No
at the ceiling and still function properly. The beam is more dispersed, and hence
c

is called broadcast. With this type of system, it is easier to line up the transceiv-
ers because of the broadcast nature of infrared light.
en
Do

TASK 4B-1:
Matching the unbounded transmission medium with
its properties
se
r

Objective: For each of the following statements, identify which

unbounded transmission media is being described.
fe
ea

1. Transmissions can’t occur through steel or load-bearing walls.

Pl
Re

2. Resistant to eavesdropping and capable of high transmission rates; how-

ever, it is susceptible to attenuation and interference.

88
3. Frequencies are divided into channels or hops.

4. Has a transmission delay of typically 240 to 300 milliseconds.

ial
5. Commonly used for long-distance voice and video transmissions, and for
short-distance high-speed links between buildings.

er
A Comparison of Media Types
The following table compares the different data-transmission media:

Twisted-pair Coaxial Fiber optic

at
cabling cable cable Unbounded media
Susceptibility to High Moderate None Low to moderate; radio is
EMI high

y
Security Low Low High Typically very low

op
Bandwidth
(capacity)
High (1-100 Mbps) Moderate
(10 Mbps)
eM
High (100
Mbps)
Varies widely

Cost to purchase Low to moderate Moderate Moderate to Moderate to high

and maintain high

tC
Ease of Easy Easy Difficult Varies widely
installation and
maintenance
Attenuation High Moderate Low Varies; affected by weather
No
and other objects
c
en

TASK 4B-2:
Do

Determining the type of media to use

Objective: For each of the following scenarios, indicate which type of
transmission media you would use, and explain why.
se
r

1. Your offices are located in an historic 19th-century building. You don’t

fe

want unsightly cables running across the floor, but you are not allowed
ea

to run cables through the walls, floors, or ceiling.

Pl
Re

2. You need to provide a high-speed connection between two offices that

are separated by a courtyard. The offices are in separate buildings.

Lesson 4: The Physical Media 89

 3. You need a high speed connection between graphic workstations that are
in the same office. You have no restrictions except the need for speed.

ial
4. You need to run a connection through a tunnel connecting two
buildings. The tunnel contains many high voltage lines and large electric
motors. You are concerned about security.

er
TOPIC 4C
Signaling

at
Signaling is how computers represent data as an electrical or light impulse
pattern. It is a function of the Physical layer of the OSI model and is specific to
the type of network you use. There are two general approaches for representing

y
data: digital signaling and analog signaling . Because of the wide variety of

op
equipment used in computer networks, you are likely to find both schemes used
eM
somewhere in the network:
• Digital devices use discrete values, such as the presence or absence of an
electrical signal, to represent portions of the data.

tC
• Analog devices use a continuous variable signal (a waveform) to represent
portions of the data.
No
Using Switches to Represent Digital Values
c

Most electronic computers use electronic switches to represent digital values. A

single switch can represent the most basic element of information, called a bit. A
en

switch has two states: off and on. The value contained in a bit corresponds to the
Do

state of the switch, 1 or 0, depending on whether the switch is on or off. To rep-

resent a number larger than 1, you have to use more than one switch. A simple
way to represent values larger than 1 is to have as many switches as the highest
number you need to represent. For example, a series of 200 switches can repre-
sent any number between 0 and 200. Even with today’s level of miniaturization,
se
r

this method would require a huge computer to perform simple tasks.

To represent and process numbers, most digital computers use the binary number
fe

system (base 2), which has only the two digits 1 and 0. In comparison, in the
ea

base 10 numbering system there are 10 digits (0 through 9). To represent numbers
greater than the number of digits available, you add another column to the left.
For example, the number greater than 9 is 10; the number greater than 99 is 100;
and so forth. This is the same scheme used in the binary system, except that you
Pl
Re

use only two digits, 1 and 0. Compare the numbers in the chart below to see how
this works:

Base 10 Base 2 (binary) Base 10 Base 2 (binary)

0 0 8 1000

90
Base 10 Base 2 (binary) Base 10 Base 2 (binary)
1 1 9 1001
2 10 10 1010
3 11 11 1011
4 100 12 1100

ial
5 101 13 1101
6 110 14 1110
7 111 15 1111

er
TASK 4C-1:
Working with binary numbers

at
Objective: To review how large decimal numbers are built in order to
gain an understanding of how binary numbers are built.

y
1. Write each of these decimal numbers in the columns in the following
table. The first number has been filled in for you.

op
eM
The Thousands Hundreds
number (103) (102) Tens (101) Ones (100)
4328 = 4 3 2 8

tC
9488 =
101 =
3=
No
90 =
c
en
Do

Using the table above, you can see that large decimal numbers are built
using the digits 0-9 and by placing those digits in columns representing
powers of ten (thus the reason these are called base-10 numbers).
se
r

2. Using the same technique, write each of these decimal numbers as

binary numbers by filling in the digits 0 and 1 into the appropriate
fe

columns. The first number has been completed for you.

ea

Decimal
number Eights (23) Fours (22) Twos (21) Ones (20)
Pl
Re

14 = 1 1 1 0
2=
12 =
7=
4=

Lesson 4: The Physical Media 91

 From the preceding table, you should see that binary numbers are built just

ial
like decimal numbers. Only the digits 0 and 1 are used. The values of the
columns are different; however, they too are powers of their base number. In
this case, each column is a power of 2.

3. How would you write the decimal number 30 using the columns of the
binary system?

er
4. How many places (columns) would you need to write the decimal num-
ber 255 as a binary number?

at
y
op
eM
Characteristics of Digital and Analog Devices
Analog electronic devices use a variable quantity (such as voltage, which is the
amount of electrical force or pressure) to represent a value. Digital devices (such

tC
as personal computers) use bits to represent values. These bits are called discrete
values.
No
c
en
Do
se
r
fe
ea
Pl

Figure 4-14: Analog waves compared to digital waves.

Re

Analog signals carry continuous values and resemble sine waves. Digital signals
carry discrete values and appear as square waves.

92
TASK 4C-2:
Identifying analog and digital devices
Objective: To determine the difference between analog and digital dis-
plays of information.

ial
1. Identify whether the following devices have analog or digital displays of
information:

er
at
y
op
eM
tC
No
c
en
Do

Figure 4-15: Digital and analog devices.

2. List at least one analog device that is not displayed above.

se
r
fe

3. List at least one digital device that is not displayed above.

ea
Pl
Re

Lesson 4: The Physical Media 93

 TOPIC 4D
Signal Transmissions
Two techniques can be used to transmit signals over the data transmission media.

ial
These are baseband and broadband transmissions. A baseband transmission line
provides only one channel per line, while a broadband transmission line provides
multiple communication channels. An example of a broadband transmission is
cable television, which uses a single coaxial cable to bring many television chan-
nels into a home.

er
at
y
op
eM
tC
No
c

Figure 4-16: Baseband and broadband signal transmission.

en
Do

Baseband
Baseband systems use digital encoding techniques to carry digital data over a
baseband: digital transmission line. Most typically, the presence of a voltage is used to rep-
A signal transmission that resent a bit with the absence of that voltage being a ″space″ between a bit.
uses all of a channel’s Because digital transmissions are affected significantly by attenuation, baseband
se
r

bandwidth. transmission lines are generally limited in length only to a kilometer or so.
In baseband transmission, all of the available frequencies in the transmission
fe

medium are used by one signal. Therefore, the medium provides only one com-
ea

munication channel, over which only one computer can successfully transmit at a
time. Baseband transmission is like two-way radio transmission: users must take
turns talking. In baseband transmission, when one computer transmits, all other
computers must be ″quiet,″ or a collision will occur and the data will be garbled.
Pl
Re

94
Broadband
Broadband systems use modulation techniques to transmit digital data over analog
carrier waves. The amplitude or frequency of the signal can be used to represent broadband:
bits, as can far more complex modulation techniques. By dividing the frequencies A transmission channel that
provided by the medium, multiple communication channels can be provided by a can carry multiple signals
single transmission line. Because broadband systems use analog signaling (which simultaneously.

ial
is less susceptible to attenuation than digital signaling), broadband transmission
lines can be much longer than baseband lines.
Broadband media are unidirectional. Unlike baseband, signals move only in one
″downstream″ direction. To enable devices to transmit to ″upstream″ computers,
there must be two data paths, one through which computers can transmit data,

er
and one through which computers can receive messages. A headend device at the
end of the transmit line forwards data to the receiver line, so that all computers
are effectively downstream of transmitted data.

at
y
op
eM
tC
No
c
en
Do

Figure 4-17: Two types of broadband systems.

As you can see in the preceding figure, dual-cable systems use two cables for the
transmit and receive data paths, while split systems use one cable with two differ-
se
r

ent frequencies for the transmit and receive data paths. The split system offers the
benefit of using a single cable, but the headend in the dual-cable system is much
fe

less complex (and usually less expensive) than the split-system headend. The
ea

split-system’s headend must perform more processing than the dual-cable system,
so the split system’s headend is called an active headend . The dual-cable sys-
tem’s headend is called a passive headend .
Pl
Re

Lesson 4: The Physical Media 95

 TASK 4D-1:
Comparing baseband and broadband
Objective: Identify the key differences between the two signal transmis-
sion schemes in the space provided.

ial
1. Baseband

er
2. Broadband

at
y
op
eM
Summary
In this lesson, you’ve examined bounded and unbounded media and the

tC
associated connectors, compared analog and digital signaling, and compared
baseband and broadband transmissions.
No
LESSON 4 REVIEW
c
en

4A Define bounded transmission media and give an example.

Do

4B Define unbounded transmission media and give an example.

se
r
fe
ea

4C What are the two general approaches for representing data? Define
each.
Pl
Re

96
4D Identify and define the two techniques that can be used to transmit sig-
nals over the data transmission media.

ial
er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 4: The Physical Media 97

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
The Data Link Layer LESSON
5

ial
Overview Data Files:
The second layer of the OSI Model is the Data Link layer. This lesson intro- none
duces the responsibilities carried out at this layer. You’ll look in-depth at the
Lesson Time:

er
MAC and LLC subdivisions of the Data Link layer. You’ll also examine
IEEE 802 specifications. 2 hours

Objectives

at
To better understand the responsibilities of the Data Link layer, you will:

y
5A Identify Data Link layer considerations.

op
eM
In this topic, you’ll identify what types of actions occur at the Data Link
layer and identify the sublayers, Media Access Control and Logical Link
Control.

tC
5B Describe the functions of the Media Access Control (MAC) and Logi-
cal Link Control (LLC) portions of the Data Link layer.
You’ll examine how MAC controls contention, polling, and token pass-
ing, and its role in resolving addresses. You’ll also see how the LLC
No
layer passes information using synchronization.
c

5C Examine the IEEE 802 specifications.

It is important that you know the specifications and limitations of various
en

cabling options. This topic introduces you to the IEEE 802 specifications
Do

and focuses on 802.2, 802.3, and 802.5.

se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 99

 TOPIC 5A
Data Link Layer Considerations
The Data Link layer uses the transmission facilities provided by the Physical

ial
layer to transfer data. Whereas the Physical layer operates on signals, the Data
Link layer operates on data. Among the responsibilities of the Data Link layer
are:
• Sharing the media among nodes; therefore, node addressing is important.
• Controlling data flow between the transmitter and receiver.

er
• Packaging data in frames. Frames add special character (bit) sequences at the
beginning or end of data to indicate the beginning and end of a unit of data
transmission, to provide reliability checks, to synchronize the transmitter and
receiver, and to perform other transmission-control functions. Large blocks
of data might need to be split up (fragmented), transmitted in several frames,

at
and reassembled at the receiver.
• Supporting the data transmission medium. Factors include topology and
media-access method.

y
• Detecting and correcting transmission errors from the Physical layer, on the

op
eM
receiving end.
For the Data Link layer to perform its functions, both the sender and receiver
must work in concert. The networking and connection devices associated with the
Data Link layer are bridges, switches, and network interface cards (NICs). Data

tC
Link-specific information, in the form of headers and trailers, enable the Data
Link layer on both ends of the transmission to communicate. Information in the
headers and trailers can consist of addresses (hardware and physical), frame
length, and upper-layer protocol information.
No
c

Among the Data Link layer considerations that you will examine are the ways in
which communication takes place, such as:
• Standards and protocols. For the Data Link layers on both the sender and
en

receiver ends to work together, they must follow the same protocols. Data
Do

Link layer protocols are often divided into two groups. The lower sub-layer
of the Data Link layer, the Media Access Control (MAC), performs Data
Link layer functions related to the Physical layer, such as controlling access
and encoding data into a valid signaling format. The upper sub-layer of the
MAC: Data Link layer, the Logical Link Control (LLC), performs Data Link layer
se

Media Access Control. The

r

functions related to the Network layer, such as providing and maintaining the
portion of the Data Link layer
that controls access to the link to the network. Many of the actual protocols functioning in these sub-
layers will be discussed in a later lesson.
fe

communication channel.
•
ea

Techniques for providing the above-mentioned functions including framing,

flow control, and error checking pertaining to the Physical layer.
LLC:
Pl

Logical Link Control.

Re

Performs Data Link layer

functions related to the
Network layer, such as
providing and maintaining
the link to the network.

100
TASK 5A-1:
Discussing Data Link layer considerations
1. Compare the operation of the Physical layer and the Data Link layer.

ial
2. Which networking and connection devices work at the Data Link layer?

er
3. What are the sublayers of the Data Link layer?

at
TOPIC 5B

y
Media Access Control (MAC) and Logical Link

op
Control (LLC)
eM
The Data Link layer is divided into two sublayers: the Media Access Control

tC
(MAC) layer and on top of that, the Logical Link Control (LLC) layer.

MAC
No
In communication systems where more than one transmitter has access to the
c

same communication channel, a standard procedure must be established to pre-

vent or handle the problem of multiple transmitters trying to send messages out
on the media at the same time. If signals are transmitted simultaneously, colli-
en

sions occur. A collision is the loss of data due to signals mixing together and
Do

becoming damaged beyond effectiveness. The Media Access Control (MAC) sub-
layer contains methods that logical topologies can use to regulate the timing of
data signals and eliminate collisions. collision:
The loss of data due to
The concept behind MAC is similar to a staff meeting in which a variety of dif- signals mixing together and
se

ferent methods can be used to ensure that everyone has an opportunity to speak. becoming damaged beyond
r

Potential problems include: effectiveness.

•
fe

Knowing when a speaker has finished speaking, thus preventing

interruptions.
ea

• Determining who has the microphone when two speakers try to speak at the
same time.
•
Pl

Determining how long a speaker can speak.

Re

• Determining speaking order.

Lesson 5: The Data Link Layer 101

 In network environments, media-access methods compensate for these problems.
Three common media-access methods are contention, polling, and token passing .
When deciding the best method for your network, keep your network’s character-
istics in mind, such as the number of network devices, quantity of transmissions,
and how quickly data needs to reach its destination.

ial
MAC Addresses
Many things on a network, such as software, hardware, and processes, have
addresses. The MAC address concerns a device’s actual physical address, which
is usually designated by the hardware manufacturer. Every device on the network
The first six places in the must have a unique MAC address to ensure proper transmission and reception of

er
MAC address on a NIC data. The MAC layer places this address on frames sent out on the network,
identify the manufacturer of
allowing the device with the matching unique address to receive the frame and
the NIC.
accept the data. ARCnet and LocalTalk LAN boards represent devices whose
MAC addresses aren’t set by the manufacturer. The installer usually sets ARCnet
addresses, and LocalTalk addresses are dynamically assigned.

at
Bridges and LAN switches, which deal with the Data Link layer, specifically fil-
ter or forward traffic based on the MAC addresses. Upon receiving frames, a

y
bridge builds tables that list network devices and the segments they reside on by
reading the MAC information in the frames; you don’t have to manually program

op
eM
the table information.

TASK 5B-1:
Discussing the MAC sublayer
tC
No
1. What is the purpose of the MAC sub-layer?
c

2. What is a collision?
en
Do

3. Describe MAC addresses.

se
r
fe
ea

LLC
Pl
Re

Logical Link Control (LLC) is the second of two portions of the Data Link layer.
The MAC and LLC sub-layers work in tandem to create a complete frame. The
portion of the frame for which LLC is responsible is called a Protocol Data Unit
(LLC PDU or PDU) . When the LLC sub-layer passes the PDU to the MAC sub-
layer for transmission, the MAC sub-layer adds the header and trailer information
that makes it a complete frame.

102
One purpose of LLC is to perform upper Data Link layer functions, such as flow
control and management of connection errors. Another is to provide the means
for transmission synchronization, which ensures that data transmissions are syn-
chronized when they travel from transmitter to receiver.
Because LLC resides in the area between the Network layer and the MAC lower
layers, it provides a standard interface through which the Network layer can com-

ial
municate with any of the lower-layer protocols that LLC supports.

er
at
y
op
eM
Figure 5-1: Communication between the LLC and the Network layer.

tC
Frame Transmission and Synchronization
On a network, bits are arranged into classes, including characters, bytes, octets,
and frames. Bits are synchronized either synchronously or asynchronously. How-
No
ever, they are synchronized at only the bit level. Synchronization must take place
c

at another level to accommodate characters, bytes, octets, and frames.

The methods by which transmissions can be synchronized are asynchronous, syn-
chronous, and isochronous. When deciding which method is best for your
en

network, keep in mind that synchronization can have a great amount of overhead,
Do

and that a method not well suited for a network that often experiences significant asynchronous:
delays in transmissions. A bit synchronization
transmission technique that
uses start and stop bits.

Asynchronous Frame Transmission

se
r

Asynchronous frame transmissions leave clock maintenance to each individual synchronous:

fe

device; transmitting and receiving devices use comparable timing, but their inter- Transmission of a bit stream
ea

nal clocks are not synchronized. A start bit is issued, followed by the frame. The of data where the transmitter
devices take no notice of whether another device has transmitted a frame. Timing and receiver are
errors can occur, but the transmission is not affected because the frames are short. synchronized.
An example of asynchronous frame transmission is a user typing on a computer
Pl
Re

keyboard.
isochronous:
Uses a single network device
for clocking and all other
devices set their internal
clocks to this one device.

Lesson 5: The Data Link Layer 103

 Parity-bit Error Detection
You can implement error-detection by adding a parity bit to every frame. A parity
bit is an additional bit sent with a frame. When the receiver adds up the character
bits and parity bit, it should get an even number if even parity error detection is
used, and an odd number if odd parity is used. If the receiver detects an error, it
can request a retransmission of the bad character.

ial
• If even parity error detection is used, then the transmitter sets the parity bit
to 1 or 0 so that the total number of ″1″ bits in the character, when added to
the parity bit, equals an even number. For example, if the character to be
sent consisted of the bits ″01101011″ (which includes five ″1″ bits), then the
transmitter would set the parity bit to ″1.″ Adding the character bits and the

er
parity bit results in an even number (6).
• If odd parity were used in this example, the receiver would have set the par-
ity bit to ″0″ so that character bits added to the parity bit would equal an
odd number (5).
Although the parity bit can successfully indicate a transmission problem that

at
occurs in a single bit, multiple-bit problems might not be discovered by this
error-detection scheme. For example, a transmission problem that changed the
bits ″01″ to ″10″ would not be detected, because the same number of ″1″ bits are

y
present.

op
eM
Typically, the even parity error-detection scheme is used in asynchronous trans-
mission; the odd, in synchronous transmission.
The transmitter and receiver must use the same parity-bit scheme. Parity-bit error

tC
detection is only about 50% efficient.

Synchronous Frame Transmission

No
c
Synchronous frame transmissions employ both the transmitter and receiver to pro-
vide transmission clocks. They do so in the form of either a separate clock line
(discussed in the Physical layer lesson) or a bit-string or a control character or
characters. Each method provides information specifying which bit begins a
en

frame or byte. The receiver uses that information to determine the point at which
Do

it should accept data and count the bits. Because the transmitter and receiver use
the same clock, timing errors are reduced. Synchronous transmissions also reduce
the overhead when transmitting large amounts of data.
During times when no data is being transmitted, fill bits are used to keep the
se

transmission timing accurate.

r

Synchronous frame transmission often employs an error-detection scheme known

fe

as Cyclic Redundancy Check.

ea
Pl
Re

104
Cyclic Redundancy Check
With the Cyclic Redundancy Check (CRC) error-detection scheme, the transmitter
performs calculations (an algorithm) on the contents of the transmission. The
transmitter attaches the result of these calculations (typically in two or more
bytes) to the data and sends it all to the receiver. The receiver performs the same (CRC) Cyclic Redundancy
calculations on the data and compares the result with the one sent by the Check:

ial
transmitter. If the two results are identical, the transmission is assumed to be An error-detection method
used to detect data
good. If not, the receiver can request a retransmission of the data. CRC is more corruption due to
efficient than parity-bit error detection; it has been said that CRC is 99% efficient. transmission errors.

Isochronous Frame Transmission

er
Isochronous frame transmission uses a single network device for clocking. All
other devices set their internal clocks to this one device, which provides continu-
ous clocking. Network devices monitor the transmission for available space
(slots) by using their media-access rules; they can insert data when there is a free

at
slot.
Unlike asynchronous systems, which provide a clock signal for every frame, and

y
synchronous systems, in which the transmitter sends a clock signal with each bit
string or character group, an isochronous system relies only on one network

op
eM
device to keep all network device clocks synchronized.
Advantages of isochronous transmission include:
• Low overhead.

tC
• Transmission rates are guaranteed; the method is deterministic.
A disadvantage is that a reliable fault-tolerant clocking device is required.
No
TASK 5B-2:
c

Discussing synchronization methods

en
Do

1. What is the difference between synchronization at the Data Link layer

and synchronization at the Physical layer?
se
r

2. In asynchronous frame transmissions, does synchronization occur? What

fe

is an example of an asynchronous frame transmission?

ea
Pl
Re

3. If the parity bit is set to ″1,″ what parity error detection is being used in
the example 11010110? What is the result?

Lesson 5: The Data Link Layer 105

 4. In the previous example, what would the result be if the parity were set
to ″0″?

5. In a CRC detection scheme, what is used to determine whether a trans-

mission is good?

ial
CHECK YOUR SKILLS 5-1

er
Suggested time:
2 minutes
Identifying synchronization methods
Identify the synchronization methods described.

at
1. Relies on only one network device to synchronize all network clocks.

y
op
2.
eM
Sends a clock signal with every bit or character group.

3. Sends a clock signal with every frame.

tC
No
Connection Services Provided by LLC
c

LLC supports the following three types of connections for transmitting data:
• Unacknowledged connectionless service. LLC supports simple point-to-point,
en

multipoint, and broadcast transmission of frames. With this type of service,

Do

LLC does not acknowledge that data has been successfully transferred; upper
layers must provide their own reliability checks. Because unacknowledged
connectionless service does not have to perform reliability checks or main-
tain a connection, it is very fast. For this reason, it is the most commonly
used type of service.
se
r

• Connection-oriented service. LLC establishes and maintains a virtual connec-

tion between two stations. This service performs flow control, sequencing,
fe

and error-recovery functions by using acknowledgments. Once the connec-

ea

tion is established, blocks of data can be transferred between nodes until one
of the nodes terminates the connection.
• Acknowledged connectionless service. LLC combines features of the other
Pl

two services. It is similar to unacknowledged connectionless service, except

Re

that it provides a mechanism through which individual frames can be

acknowledged.
An acknowledgment is simply a message that confirms a frame’s reception.

106
 ial
er
at
y
op
Figure 5-2: Types of services provided by LLC.
eM
tC
TASK 5B-3:
Describing LLC connection types
No
1. Describe the three types of connections provided by LLC.
c
en
Do
se
r

2. Identify how a network that uses unacknowledged connectionless service

fe

might cope with transmission errors.

ea
Pl
Re

Flow-control Techniques
Flow-control techniques ensure that data is not transmitted faster than the receiver
can process it. Two forms of flow control include:

Lesson 5: The Data Link Layer 107

 • Window flow control:
— sliding-window (dynamic)
— static-window
• Stop-and-wait flow control (guaranteed rate)

ial
er
Figure 5-3: Sliding-window flow control.

Sliding-window Flow Control (Dynamic)

at
Some networks, such as those using satellite microwave links or large
internetworks, experience a significant delay between the time a transmission is

y
sent and the time it is received. In such networks, the long delays involved in
waiting for acknowledgment can be unacceptable. To overcome this problem, the

op
eM
sliding-window flow control technique does not acknowledge each frame
individually. Instead, several frames are sent as a stream, the size of which varies.
As the transmitter sends the stream, the receiver’s buffer fills. When the incoming
stream surpasses the receiver’s buffer size, a choke packet is sent back; it tells the

tC
transmitter to slow down. The transmitter complies, but then slowly increases the
stream again until the receiver’s buffer fills and the receiver sends another choke
packet. In this way, the data stream size constantly varies. Sliding-window is also
referred to as dynamic or floating-window flow control.
No
c
en
Do

Figure 5-4: Static window flow control.

se
r

Static-window Flow Control

fe
ea

The protocols that use static-window flow control use a specific window size. For
example, a window size may be equal to seven frames. Every frame in the trans-
mission is assigned a number from 1 to 7. An acknowledgment is sent for each
frame. Another frame cannot be sent until a frame with that same number assign-
Pl
Re

ment has been acknowledged. This scheme allows the maximum number of
outstanding frames to be equal to the window size; in this example, the size of
the window is seven frames.

108
Stop-and-wait Flow Control (Guaranteed)
In protocols that use stop-and-wait flow control, the transmitter sends a frame and
waits for the receiver to acknowledge its receipt. The transmitter cannot send
another frame until the previous frame has been acknowledged.
Stop-and-wait flow control protocols are sometimes called reliable, or guaranteed

ial
flow control, because the transmitter receives acknowledgment when data has
been received at the destination. This type of control is established before trans-
missions are sent.

er
at
Figure 5-5: Stop-and-wait flow control.

y
Error Control

op
eM
In the LLC sub-layer, error control consists of being informed of lost or garbled
frames. The following table describes instances that will invoke LLC error
control.

Instance
Frame fails to reach receiver
Explanation
In a connection-oriented or acknowledged connectionless
service, if a negative acknowledgment (NAK) or no
tC
No
acknowledgment is returned to the transmitter, an error is
c

presumed and retransmission occurs.

Offending frame size A receiver will not accept a frame that is higher than the
maximum size limit or lower than the minimum size limit.
en

Differing checksums If a receiver’s checksum differs from the transmitter’s, an error

Do

results. The receiver can request retransmission.

Capacity limit If a channel or device is holding all that it is capable of, any
subsequent frame transmissions can result in an error.
Interference Interference, such as noise, can garble or lose data, resulting
in a transmission error.
se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 109

 TASK 5B-4:
Discussing flow-control techniques
1. Describe the use of sliding-window flow control.

ial
2. Describe the use of static-window flow control.

er
at
3. Describe the use of stop-and-wait flow control.

y
op
4.
eM
List some instances that will invoke LLC error control.

TOPIC 5C tC
No
IEEE 802 Specifications
c

In 1980, the Institute of Electrical and Electronics Engineers (IEEE) met to

develop a set of standards for network cabling, topologies, and media-access
en

schemes (Physical and Data Link layer standards). They convened in February,
Do

The term IEE 802 1980, so the standards that they developed were designated ″802.″ The IEEE
specifications is derived from assigned Technical Advisory Groups to work on various sections of the 802
the fact that the IEEE
project. Each section is related to a particular domain of data communications, as
committee met in 1980 (80)
in February (2), thus 802. shown here.
se
r

Section Domain covered

IEEE 802.1 Communication between two IEEE 802 stations on different WANs or LANs.
fe

IEEE 802.2 Logical Link Control (LLC) for 802.3, 802.4, 802.5, and 802.6
ea

implementations.
IEEE 802.3 Contention-based standards.
IEEE 802.4 Token-passing standards for bus topologies.
Pl
Re

IEEE 802.5 Token-passing standards for star or ring topologies.

IEEE 802.6 Distributed Queue Dual Bus standard for municipal area networks.
IEEE 802.7 Installation and maintenance standards for broadband networks.
IEEE 802.8 Optical-fiber standards.
IEEE 802.9 Isochronous Ethernet standards for ISDN-integrating voice and data
networks for time-sensitive communication.

110
Section Domain covered
IEEE 802.10 Network security or VLANs.
IEEE 802.11 Wireless LANs.
IEEE 802.12 100Mbps contention-based network.

ial
These standards were adopted by ANSI (American National Standards Institute),
and then later revised by the ISO. They are sometimes known as the IEEE 802.x
series. A relationship between each protocol and OSI model layers exists, as
shown in Figure 5-6. The following topics in this lesson provide details about the
IEEE standards listed in the table above.

er
at
y
op
eM
tC
Figure 5-6: IEEE 802.2 and the OSI model.
IEEE 802.2 defines the Logical Link Control (LLC) standard that performs func-
tions in the upper portion of the Data Link layer, such as flow control and
No
management of connection errors. The lower portion of the Data Link layer
Remember that 802.2 works
c

(Media Access Control, or MAC) is provided through other standards, such as

IEEE 802.5 and IEEE 802.3. at the Data Link layer of the
OSI model.
Because LLC resides in the area between the Network layer and the lower-layer
en

IEEE standards (802.3, 802.4, and 802.5), it provides a standard interface through
Do

which the Network layer can communicate with any of the lower-layer protocols
that LLC supports. IEEE 802.2 identifies a frame’s upper-layer protocols and the
Network layer processes used by the frame’s source and destination by adding
header fields.
IEEE 802.2 provides a large degree of flexibility, making it possible for a wide
se
r

variety of network media to be interconnected on the same network.

fe
ea

802.3 and Ethernet

Ethernet, which was first available in the late 1970s, is still an accepted standard
Pl

because of its respectable bandwidth, economical price, and compatibility with

Re

other network products.

Lesson 5: The Data Link Layer 111

 IEEE 802.3 is an extension of the original Ethernet. Because IEEE 802.3 is a
form of Ethernet, it is often called ″Ethernet″ instead of ″IEEE 802.3.″ However,
IEEE 802.3 includes modifications to the classic Ethernet data-packet structure.
Furthermore, while Ethernet spans the Physical and the Data Link layers, IEEE
802.3 spans the Physical layer and only the lower portion of the Data Link layer;
IEEE 802.2 provides the functions of the upper portion of the Data Link layer.

ial
IEEE 802.3 and the Physical Layer
802.3 specifies several Physical layer configuration options, such as baseband and
broadband transmissions, several types of media and topologies, and data rates.
Each configuration option has the same thing in common—they use the Carrier

er
Sense Multiple Access/Collision Detection (CSMA/CD) media-access method.
802.3 supports five different implementation configurations: 10BASE5; 10BASE2
(sometimes called thin-net or cheaper-net); 1BASE5 (also known as Star LAN);
10BASET; and 10BROAD36.

at
The configuration names indicate the following three IEEE 802.3 characteristics:
• Data rate (megabits per second). The first number in the configuration name.

y
• Type of signaling (baseband or broadband). The middle portion of the name.
• Maximum segment length (rounded to the nearest 100 meters). The last num-

op
eM
ber in the name. ( T indicates that the configuration uses twisted-pair cable.)

tC
IEEE 802.3 Specifications
For comparison, the following table describes various IEEE 802.3 media formats,
including their data rates, topologies, and segment lengths. The maximum dis-
tance between stations is 500 meters.
No
c

Data Rate (in Maximum segment

Medium Mbps) Topology length (in meters)
en

10BASE5 (thick coax) 10 Bus 500

Do

10BASE2 (thin coax) 10 Bus 185

1BASE5 (UTP) 1 Star 250
10BASE-T (UTP) 10 Star 100
10BASE-F (optical fiber) 10 Star 500-2000
100BASE-T (twisted pair) 100 Bus 4000
se
r

10BROAD36 (broadband 10 Bus 3600

coax)
fe
ea
Pl
Re

112
TASK 5C-1:
Choosing an IEEE 802.3 configuration
Objective: Use the IEEE 802.3 specifications table to identify which
IEEE 802.3 configuration you would use in each of the fol-

ial
lowing situations, and explain why.

1. You need to connect several offices in a large industrial complex that

spans more than half a mile.

er
2. You need to install a small LAN in a single office. Your budget limits
how much you can spend.

at
3. You plan to install LAN cabling above the ceiling tiles in your office.

y
There are other electrical lines and fan motors above the ceiling tiles,
and the cables must be run through several tight curves.

op
eM
Collision Detection
IEEE 802.3 employs a contention-based media access method, enabling networktC
devices to transmit data at will. When two devices attempt to transmit at the
No
same time, 802.3’s collision detection process kicks in to stop that from happen-
c

ing, as shown in the following figure.

en
Do
se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 113

 ial
er
at
y
op
eM
Figure 5-7: IEEE 802.3’s collision detection process.

Differences Between IEEE 802.3 and Ethernet

• tC
The following is a summary of the differences between IEEE 802.3 and Ethernet:
Ethernet supports the Physical and Data Link layers, while IEEE 802.3 sup-
ports the Physical layer and only the lower half (MAC sublayer) of the Data
No
Link layer. IEEE 802.2 supports the upper half (LLC) of the Data Link
c

layer.
• Ethernet supports only one data rate, transmission medium, and topology.
IEEE 802.3 supports a variety of data rates, transmission media, and
en
Do

topologies.
If you compare Ethernet specifications with those of IEEE 802.3, you will see
that 10BASE5 is the IEEE 802.3 configuration that is most like Ethernet.

Ethernet specifications
se
r

Data Rate (in megabits/second) 10

Maximum distance between stations (in meters) 500
fe

Maximum segment length (in meters) 500

ea

Type of media Thick or thin coax, UTP, and duplex optical fiber
Media Access Control Contention-based CSMA/CD
Topology Bus
Pl
Re

Signaling technique (baseband) Manchester

114
Ethernet Media Types
Next, let’s look at each of the Ethernet media types in more detail.

Thick Coax Cable

ial
10BASE5, also called Thicknet or Thick Ethernet, uses thick, coaxial cable. As in
other types of networks, 802.3 devices attach to the network through network
adapters. 802.3 network adapters typically have an onboard 802.3 transceiver and
use an 802.3 T-connector to tap into 802.3 thin Ethernet cable. Alternatively, they
use an 802.3 Attachment Unit Interface (AUI) cable to attach to a separate trans-
ceiver called an 802.3 Medium Attachment Unit (MAU), which is attached

er
directly to the 802.3 thick Ethernet cable transmission line.
Thick coax cable (RG-6) requires the following:
• a 50-ohm terminator on each end of the cable;
• a maximum of 3 segments with attached devices (populated segments);

at
• a network board using the external transceiver;
• a maximum of 100 devices on a segment, including repeaters;

y
• a maximum length of 1,640 feet (500 meters) per segment;

op
•
eM
a maximum of 4,920 feet (1500 meters) per segment trunk;
• one ground per segment;
• a maximum of 16 feet (5 meters) between a tap and its device; and

tC
• a minimum of 8 feet (2.5 meters) between taps.
Thick coax cable is inflexible which can make changes, such as adding or mov-
ing workstations and devices, difficult. It also has some distance restrictions.
No
c

Thin Coax Cable

Thin coax cable, or Thin Ethernet, implemented with T-connectors and termina-
en

tors, such as RG-58 and A/U or C/U, have the following specifications:
Do

• a 50-ohm terminator on each end of the cable;

• a maximum length of 1,000 feet (185 meters) per segment;
• a maximum of 30 devices per segment;
•
se

a network board using the internal transceiver;

r

• a maximum of 3 segments with attached devices (populated segments);

•
fe

one ground per segment;

ea

• a minimum of 1.5 feet (.5 meters) between T-connectors;

• a maximum of 1,818 feet (555 meters) per trunk segment; and
• a maximum of 30 connections per segment.
Pl
Re

Lesson 5: The Data Link Layer 115

 Thin Ethernet cable is flexible; however, bending or twisting it can damage or
break it. If it is broken, the whole segment fails. A thin coax implementation can-
not use drop cables, and requires two coax cables to come from a wall in every
Two pieces of thin coax person’s office, which can be a nuisance. Also, the more devices you have, the
cables come into each harder it is to make changes to your network layout. 10BASE2 uses thin Ethernet
person’s office or cubicle;
cable.
one goes to each neighbor’s

ial
office as the coax makes its
way around the office.
Twisted-pair Ethernet Cable
Twisted-pair Ethernet cable has the following specifications:
• a maximum of 1,024 attached workstations;

er
• a maximum of 4 repeaters between communicating workstations; and
• a maximum segment length of 328 feet (100 meters).
Devices that use an unshielded twisted-pair implementation generally connect the
cable to a wall jack. From there, permanent cable drops connect to a punch-down

at
block, which, in turn, connects to a patch panel. Hubs combine and repeat the
signals, and can cascade to facilitate large numbers of connections.

y
op
eM
tC
No
c

Figure 5-8: Twisted-pair Ethernet cable.

UTP configurations employ a hierarchical star topology. They are easily modified
en

and have few limitations.

Do

The 5-4-3 Rule

You can use the 5-4-3 rule when designing your coax or UTP Ethernet
se

configurations. It is a network design guideline that can help you establish an

r

efficient Ethernet network.

Using internetworking
fe

devices, such as routers, you Coaxial implementation: five cable segments per network, connected by up to
ea

can overcome limitations four repeaters, allowing three populated segments.

such as the maximum
network length. UTP implementation: five cable segments per network, connected by up to four
hubs, or multiport repeaters.
Pl
Re

Note: A UTP implementation can be referred to as the 5-4 rule because UTP seg-
ments cannot have attached devices.

The following tables list the IEEE 802.3 specifications:

116
IEEE 802.3
specifications 10Base2 10BaseT3 10Base5
Data rate 10 Mbps 10 Mbps 10 Mbps
Maximum segment length 185 meters 100 meters (distance 500 meters
from hub to
computer)

ial
Minimum distance between .5 meter N/A 2.5 meters
stations
Maximum distance 500 meters N/A 500 meters
between stations
Maximum total network 925 meters 500 meters per 2460 meters

er
length collision domain
Maximum connected 5 (using 4 repeaters) 5 (using 4 hubs in a 5 (using 4 repeaters)
segments with only 3 segments cascading with only 3 segments
allowed to have configuration) allowed to have
computers connected1 computers
connected1

at
Maximum computers per 30 with a further limit 12 per hub with a 100
segment of 1024 per network further limit of 1024
per network

y
Media type Thinnet Twisted pair Thicknet

op
Signaling Baseband
eM
Baseband Baseband
Media Access scheme CSMA/CD CSMA/CD CSMA/CD
Topology Bus Star-wired bus Bus

tC
Connector to network BNC T-adapter RJ-45 DIX or AUI
adapter
Impedance 50 ±2 Ω (Ohms) 85-115 Ω with UTP 50 ±2 Ω
or 135-165 Ω with
STP
No
Terminator resistance 50 Ω (Ohms) 2
50 Ω
c

Footnotes:
en

1: This relationship is often called the ″5-4-3 rule.″

Do

2: Twisted-pair segments are terminated; however, you do not need to worry

about implementing this termination. Hubs provide internal termination for
twisted-pair networks when necessary.
3: The fiber-optic variation of 10BaseT is called 10BaseF, which includes
se
r

three sub-types: 10BaseFL (fiber link), 10BaseFP (fiber passive), and

10BaseFB (fiber backbone).
fe

IEEE 802.3
ea

specifications 100BaseVG
(continued) 100BaseT3 AnyLAN4 1Base5 10Broad36
Data rate 100 Mbps 100 Mbps 1 Mbps 10 Mbps
Pl
Re

Lesson 5: The Data Link Layer 117

 IEEE 802.3
specifications 100BaseVG
(continued) 100BaseT3 AnyLAN4 1Base5 10Broad36
Maximum segment length 2000 meters 100-200 500 meters 3600 meters
(for 100BaseFX) meters for (distance from
less for other twisted pair, hub to

ial
sub-types 2000 meters computer)
for fiber optic
(distance from
hub to
computer)
Maximum distance 205 meters (for N/A N/A 500 meters

er
between stations 100BaseTx and
T4)
Maximum total network N/A 500-1000
length meters for
twisted pair,

at
10,000 (10
km) for fiber
optic

y
Maximum connected N/A 5
segments

op
Media type
eM Twisted pair or Twisted pair Twisted pair Thicknet
fiber optic and fiber optic (broadband)
Signaling Baseband Baseband Baseband Broadband
Media Access scheme CSMA/CD Demand CSMA/CD CSMA/CD

tC
Priority
Topology Star-wired bus Star-wired bus Bus Bus
Connector to network RJ-45 RJ-45 RJ-45 DIX or AUI
adapter
No
Impedance 50 ±2 Ω
c

Terminator resistance 2 2 2
50 Ω
en

Footnotes:
Do

1: This relationship is often called the ″5-4-3 rule.″

2: Twisted-pair segments are terminated; however, you do not need to worry
about implementing this termination. Hubs provide internal termination for
twisted-pair networks when necessary.
se
r

3: The 100BaseT specification has at least three sub-types: 100BaseTx runs

over two-pair Category 5 cabling; 100BaseT4 runs over four-pair Category
fe

3, 4, or 5 cabling; and 100BaseFx runs over fiber-optic cabling.

ea

4: 100BaseVG is a technology developed by Hewlett-Packard. It was stan-

dardized by the IEEE as 802.12. The ″VG″ in the name stands for ″voice
grade.″
Pl
Re

118
TASK 5C-2:
Distinguishing between different Ethernet media
types
Objective: Determine which Ethernet type, thick, thin, or twisted pair, is

ial
appropriate for each of the following circumstances.

1. 900 feet between segments; 2 feet between T-connectors; 3 populated

segments; and 25 connections per segment.

er
2. 300 feet between segments; 350 total connections; star topology; and 2
repeaters between workstations.

at
3. 1,000 feet between segments; 2 populated segments; 100 devices per seg-
ment; and 2 repeaters between workstations.

y
op
4.
eM
Which type of Ethernet is flexible? How flexible is it?

CHECK YOUR SKILLS 5-2 tC Suggested time:

No
2 minutes
c

Using the 5-4-3 rule

Using the 5-4-3 rule, determine the appropriate number.
en
Do

1. How many repeaters can a coaxial implementation have?

2. How many populated segments?

se
r

3. How many cable segments per network?

fe
ea
Pl
Re

Lesson 5: The Data Link Layer 119

 Large networks incur frequent problems with device access and other bottlenecks
because of standard Ethernet’s 10 Mbps limitation on information transfer. New
technologies that are bringing higher CPU performance, high-speed buses, and
high-end applications insist on additional bandwidth. To address this issue, the
Fast Ethernet Alliance has devised IEEE 802.3u Fast Ethernet specifications, such

ial
as physical and logical topologies, Auto Negotiation (AUTONEG), Media Inde-
pendent Interface (MII), and Media Access Control (MAC). These specifications
are summarized in the following table.

Fast Ethernet specification Explanation

er
Topologies Physical—uses a hierarchical star configuration consisting of
individual cables connecting at a common point. Common points
interconnect by using repeaters, hubs, and switches. Logical—
uses a bus configuration.
MII Enables Fast Ethernet to use either Physical Layer Devices
(PLDs), or transceivers (100BASE-TX, 100BASE-T4, or 100BASE-

at
FX). A Fast Ethernet network board can connect to an external
transceiver with UTP cables, creating a flexible environment with
both optical fiber and copper media. Fast Ethernet specifies a

y
maximum copper segment length of 100 meters. However, the
maximum cable length between repeaters is 5 meters. A Fast

op
eM Ethernet network defined by 100BASE-FX and in half-duplex
mode has a maximum segment length of 412 meters without
repeaters—full duplex mode has a maximum length of 2,000
meters.

tC
AUTONEG During a transition to a Fast Ethernet environment, 10/100 Fast
Ethernet adapters can be installed on workstations to establish
(negotiate) a data rate in the mixed environment.
MAC No changes were made to the CSMA/CD protocol.
No
c

Fast Ethernet Physical Specifications

en
Do

Let’s examine some of the physical specifications for various cabling and repeat-
ers used in Fast Ethernet implementations.

100BASE-TX
se
r

The 100BASE-TX specification uses two pairs of Category 5 UTP or Category 1

fe

STP cabling at a 100 Mbps data transmission speed. Each segment can be up to
ea

100 meters long. UTP cables use standard RJ-45 connectors; STP cables use
DB-9 connectors.
Pl
Re

100BASE-T4
The 100BASE-T4 specification uses four pairs of Category 3, 4, or 5 UTP
cabling at a 100 Mbps data transmission speed with standard RJ-45 connectors.
Each segment can be up to 100 meters long.

120
100BASE-FX
The 100BASE-FX specification uses two-strand 62.5/125 micron multi- or single-
mode fiber media. Half-duplex, multi-mode fiber media has a maximum segment
length of 412 meters. Full-duplex, single-mode fiber media has a maximum seg-
ment length of 10,000 meters. 100BASE-FX includes the use of Media Interface
Connectors (MIC) or Subscriber Connectors (SC) specified by ANSI FDDI.

ial
Repeaters
There are two types of Fast Ethernet repeaters, classified by performance and fea-
tures: Class I and Class II. Both are 100 Mbps repeaters. One Class I repeater or

er
two Class II repeaters can be used in a single collision domain.
IEEE 802.3u repeaters sometime have to translate frequencies when interconnect-
ing dissimilar 100BASE segments. For example, 100BASE-TX and 100BASE-FX
both use a 41.6 MHz signal frequency, but 100BASE-T4 uses a 25 MHz signal

at
frequency. The repeater must translate the frequencies to connect the segments
that use different frequencies. The maximum number of repeaters per collision
domain is two with a maximum inter-repeater link cable length of 5 meters.

y
Note: A repeater that translates frequencies brings more latency to a network than

op
eM
does a single frequency repeater. Latency is the amount of time it takes a packet to
travel from sender to receiver.

TASK 5C-3:
Discussing Fast Ethernet
tC
No
latency:
c

1. Why was Fast Ethernet developed? The amount of time it takes a

packet to travel from sender
to receiver.
en
Do

2. What type of media does each of the Fast Ethernet physical specifica-
tions require?
se
r
fe

3. True or false: A repeater that translates frequencies increases network

ea

latency.
Pl
Re

802.5 and Token Ring

IBM Token Ring is essentially a subset of IEEE 802.5. Because IBM has pro-
vided attractive mainframe-to-LAN connectivity solutions for Token Ring
networks, Token Ring has become a popular lower-layer standard.

Lesson 5: The Data Link Layer 121

 Like 802.3, 802.5 provides MAC sublayer and Physical layer functions by using
802.2’s link control functions

ial
er
at
y
op
eM
Figure 5-9: An IEEE 802.5 token-passing ring. Station A transmits a frame to Station C.

802.5 Media Access Control Sublayer tC

No
IEEE 802.5 uses token passing to control access to the medium. IEEE 802.5
includes a priority scheme that enables specific nodes to control the token (and
c

access to the medium) longer than other nodes.

en
Do

Active Monitor
IEEE 802.5 requires that there be a token somewhere on the network at all times
and that stations relinquish the token when they have finished using it. To enforce
these requirements, one of the stations on the network functions as an active
monitor; any station can be the active monitor. If no active monitor exists, a sta-
se
r

tion is automatically designated.

Suppose that the station holding the token fails (loses power, for example) before
fe

it can transmit a data frame or pass the token on to the next station. Stations can-
ea

not transmit until they have the token, so no station would be able to transmit.
The same situation might occur if the token were damaged in transmission.
The active monitor constantly examines the network. If a certain interval passes
Pl
Re

and the active monitor does not detect any frames (the token or a data frame) on
the network, then the active monitor determines that the token has been lost and
generates a new token.

122
Persistently Busy Token
After a station transmits a data frame, that frame travels around the ring, eventu-
ally returning to the station that transmitted it. At that point, the station
relinquishes the token. While the token is in the possession of a station, it is said
to be ″busy.″

ial
Consider what would happen if the station holding the token failed just after it
transmitted a frame. The station would not be able to release the token to the
next station, and the data frame would travel around the network endlessly unless
it was stopped. In this case, the active monitor would mark the data frame by
setting its monitor bit (in the frame’s Access Control field). If the marked frame
were to return to the active monitor, the active monitor would assume that there

er
was a problem. It would remove the data frame from the network and pass a new
token to the next station.
Stations other than the active monitor function as passive monitors; they monitor
the activities of the active monitor. If the active monitor fails, one of the passive
monitors takes its place.

at
Priority Control

y
An optional feature called priority control enables stations to have channel-access

op
eM
priority over other stations. This is implemented through three priority bits and
three reservation bits in the Access Control field. Stations use the reservation bits
to reserve tokens. The priority of the frame being transmitted is held in the prior-
ity bits.

TASK 5C-4: tC
No
c

Discussing 802.5’s MAC sublayer

1. What two things does IEEE 802.5 require?
en
Do

2. What happens if the token is not passed to the next node?

se
r

3. What is the active monitor’s function?

fe
ea

4. True or false: It is not possible to prioritize node access to the network.

Pl
Re

Lesson 5: The Data Link Layer 123

 802.5 and the Physical Layer
Although IEEE 802.5 does not specify a particular topology, IBM’s Token Ring
has established the ring topology as the most popular. Token Ring stations con-
nect to a Multi-Station Access Unit (MSAU). Relay switches inside the MSAU
bypass ports to which no station or a defective station is connected to provide a
continuous ring. Because the ring path is hidden inside the MSAU, Token Ring

ial
networks look like star topologies. Because of this, Token Ring networks are
sometimes described as a ″star-wired ring.″

er
at
y
op
eM
tC
Figure 5-10: Because the ring ″path″ is hidden inside the MSAU, Token Ring’s topology
appears more like a star than a ring.
No
Adapter cables are used to connect MSAUs in the main ring path. By using patch
c

cables to connect MSAU’s ″ring in″ and ″ring out″ ports, you can link several
MSAUs to provide connections for more stations than a single MSAU permits.
en
Do
se
r
fe
ea
Pl
Re

Figure 5-11: Two MSAUs connected with a patch cable to enable more nodes to connect.

124
802.5 Token Ring specifications
Lobe length (distance from node to 330 ft (100 m) for Type 1 or 2 cable; 220 ft (66m) for Type 6
MSAU) or 9; 150 ft (45 m) for Type 3 UTP
Distance between nodes 8 ft (2.5 m) minimum
Ring length (distance between MSAUs 660 ft (200 m) for Type 3 cable; 400 ft (45m) for Type 6;6
on main path) mile (1 km) for optical fiber

ial
Maximum number of nodes 250 (IEEE 802.5); 260 (IBM STP); 72 (IBM UTP)
Maximum number of segments 3
between repeaters
Maximum number of MSAUs 33
Type of media None specified

er
Media access method Token passing
Data rate (in megabits/second) 4 or 16 (16 requires Type 4 or better cable)

In addition to the specifications listed above, Token Ring requires that each seg-

at
ment is terminated at both ends and grounded at one.

Be sure not to confuse 802.

5, the Token Ring

y
Beacon Frames specification, with 802.4, the
Token Bus specification.

op
eM
IEEE 802.5 includes a feature that helps the network to automatically recover
from errors, such as breaks in the network cabling. Stations detecting a serious
network error continuously transmit beacon frames around the ring until they
receive one from a station preceding them. A beacon frame enables other stations

tC
to isolate the area in which the problem occurred (the failure domain). Once a
station receives a beacon frame from a station preceding it, it stops sending bea-
con frames. When only one station is left sending beacon frames, the problem is
isolated to the station preceding the station sending frames. The station still send-
No
ing beacon frames sends a beacon MAC frame containing its addresses, its
c

nearest active upstream neighbor (NAUN), and all stations in between that could
have failed.
The network can be automatically reconfigured to remove the failure domain
en

(which includes the station that transmitted the beacon frame, the station preced-
Do

ing it on the network, and the medium between them) from the network.
The following table lists the IEEE 802.5 and Token Ring specifications.

IEEE 802.5 and Token

se

Ring specifications
r

IEEE 802.5 Token Ring

fe

Data rate 1 or 4 Mbps 4 or 16 Mbps

ea

Minimum distance between not specified 2.5 meters

stations and MSAU
Maximum distance between not specified 101 meters (with Type 1 cables),
stations and MSAU (lobe cable 100 meters (with STP), or 46
Pl
Re

distance max.) meters (with UTP)

Maximum distance between not specified 152 meters (500 ft)
MSAUs
Maximum ring length not specified 160 meters (16 Mbps over Type
1 cables) to 360 meters (4 Mbps
over Type 1 cables)

Lesson 5: The Data Link Layer 125

 IEEE 802.5 and Token
Ring specifications
IEEE 802.5 Token Ring
Maximum connected segments not specified 33 MSAUs (9 over Type 3 cable)
Maximum number of stations 250 260 with STP or 72 with UTP

ial
Media type UTP or STP UTP or STP
Signaling Baseband Baseband
Media Access scheme Token passing Token passing
Topology None specified, though Star-wired ring
typically a star-wired ring
Impedance 100-120 Ω UTP 150 Ω STP 100-120 Ω UTP 150 Ω STP

er
Terminator resistance N/A N/A
Connector to network adapter not specified Media Interface Cable (MIC) (an
IBM proprietary ″genderless″
connector), RJ-45, or RJ-11

at
y
TASK 5C-5:

op
eM
Discussing 802.5’s Physical layer
1. What is the most widely used physical topology for IEEE 802.5?

2. tC
Why do 802.5 token ring topologies look like star topologies?
No
c

3. What is the purpose of a beacon frame?

en
Do
se

Summary
r

In this lesson, you’ve examined the Data Link layer and the two subdivi-
fe

sions of the layer, MAC and LLC. You also examined IEE 802 standards for
ea

802.2, 802.3, and 802.5.

Pl
Re

126
LESSON 5 REVIEW
5A List some of the functions of the Data Link layer.

ial
5B What is the purpose of the MAC layer?

Describe how communication is handled with the upper layers.

er
Describe LLC’s connection services.

at
y
op
eM
5C Provide a general description of IEEE 802.3, and describe the differ-
ences between IEEE 802.3 and Ethernet.
tC
No
c
en

List the specifications of IEEE 802.5, and describe the difference

Do

between IEEE 802.5 and Token Ring.

se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 127

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Data Transmission LESSON
6

ial
Overview Data Files:
In this lesson, you’ll examine the hardware used in data transmission (the Network+ Certifiication
Physical layer of the OSI Model). Then you’ll examine how that data is put CD-ROM

er
on the wire (the Data Link layer of the OSI Model).
Lesson Time:
1 hour, 30 minutes

Objectives

at
To examine how data is placed on the data-transmission media, a function pro-
vided by the Physical and Data Link layers of the OSI model, you will:

y
6A Examine network connection devices.

op
eM
The purpose of network connection devices and issues associated with
them is covered in this topic.

6B Identify the role of network adapters in data transmissions.

tC
You’ll examine several bus types and the associated network adapters for
each. You’ll also look at what the network adapter is responsible for in
data transmissions. Finally, you’ll examine how to install, configure, and
troubleshoot network adapters.
No
c

6C Examine media-access control methods.

In this topic, you’ll examine contention, polling, token passing, and
demand priority media access control methods.
en
Do

6D Identify the characteristics of common network topologies.

This topic revisits network topologies and examines the advantages and
disadvantages of bus, ring, star, mesh, and hybrid topologies.

6E Describe different types of bridges and how they route data through a
se
r

network.
In this topic, you’ll examine transparent, source-routing, and source-
fe

routing transparent bridges. You’ll learn about the function provided by

ea

bridges and how they affect network traffic.

Pl
Re

Lesson 6: Data Transmission 129

 TOPIC 6A
Introduction to Network Connection Devices
Network connection devices connect the various segments of network media

ial
(cables, optical fiber, radio waves, and so forth) that make up a network or
internetwork. A segment is one section of network transmission media that is
Examples of concentrators assigned a specific network address; it provides servers and clients on the net-
include hubs, MSAUs, and work with access to network resources.
switches. Each of these
devices enables several When connecting several media segments to form a large network, you can opt to
devices to connect to it, thus use one or more network connection devices. A network connection device deter-

er
concentrating access of mines whether connected segments are on the same or different networks by
multiple devices to a single
device.
assigning them the same or different network addresses.
Several types of network connection devices exist, including repeaters, bridges,
switches, and routers . These are general names; when attempting to purchase one

at
of these devices, you might find it under another name. For example, repeaters
might also be called hubs or concentrators . There are also some less common
network connection devices, such as brouters (a bridge/router hybrid) and gate-

y
ways . Each type of device solves a specific set of network connection problems.
concentrator:

op
A general term for hubs,
MSAUs, and switches, all of
eM
which concentrate access of
multiple devices to a single
device.

tC
No
c
en
Do
se
r
fe
ea

Figure 6-1: Network connection devices manage network traffic problems.

Pl
Re

Connection Device Hardware

For devices to communicate and be networked, they must be attached to trans-
mission media. Connection hardware, such as network interface boards,
transmission media connectors, and modems, join computer devices to media
cable.

130
 ial
er
at
y
op
Figure 6-2: Network interface boards.
eM
tC
Network Interface Board
A network interface board attaches a device, such as a computer, to a transmis-
sion medium. It provides the necessary circuitry to make the connection
No
(physically and logically). The majority of the board is installed in the computer;
c

however, one end of it remains outside the computer so that the transmission
media’s connector can be attached to it, as shown in Figure 6-2. Every computer
on a network must have its own network interface board.
en

Types of network interface boards include:

Do

• Network interface cards (NIC)

• Certain external devices (transceivers)
• Software with a generic hardware port
se
r

Transceiver
fe
ea

The transceiver is the basic component of a NIC, in that this is what enables the
NIC to send and receive data.
Used by most network interface boards, transceivers send and receive EM or
electric signals over the transmission media. They are physically connected to the
Pl
Re

transmission media by a connector. Transceivers for wireless media are simply

sending and receiving devices for which no mechanical connectors are necessary.
A transceiver is often a separate box connected to a ″pigtail″ cable.

Lesson 6: Data Transmission 131

 If you need to connect dissimilar media without changing the access method, you
can use a transceiver. For example, if you need to connect a 10BaseT segment to
a 10Base5 backbone, you could use a transceiver since they both use Ethernet.
Transceivers are most often found in 10Base5 environments to connect worksta-
tions to the 10Base5 backbone.

ial
Network Interface Card
Also known as network adapters, network interface cards (NICs) are circuit
boards that provide the means to change a computer’s electric signals to those
used by the transmission medium. Sometimes a computer does not provide a net-
work port. A NIC is used to add this capability to a computer. NICs can contain

er
one or more types of media connectors, as shown in Figure 6-3.
Typically, a NIC uses an internal transceiver built into its circuitry. However,
external transceivers can be attached to the NIC’s cable or media connector, if
necessary.

at
These days, ″cards″ generally refers to PC or PCMCIA cards, like those used in
notebook PCs, which enable a computer to make a network connection. NIC is
an old expression that is synonymous with network interface board.

y
op
eM
Transmission Media Adapter
Sometimes a network interface board uses a connector different from those

tC
already in use by the transmission medium. When this happens, the signals from
one connector can’t communicate with those from the other connector. To prevent
this, a transmission media adapter is used. A transmission media adapter receives
signals from one of the connectors and converts them to a format interpretable by
the other connector.
No
transmission media
c

adapter:
Receives signals from one of
the connectors and converts
en

them to a form interpretable

by the other connector.
Do
se
r
fe
ea
Pl
Re

132
 ial
er
at
y
op
eM
tC
Figure 6-3: Various transmission media connectors.

Transmission Media Connector

No
c

A transmission media connector joins a network device to a media cable. Each

media type has a compatible transmission connector (see Figure 6-3). Some
media have more than one from which you can choose. Media connectors attach
en

to media cable through the network interface board. transmission media

Do

connector:
Joins a network device to a
media cable.
Modem
Computers use electric pulses to send data signals, but some media, such as
se

microwave, as well as public phone networks, use EM waves. This presents a

r

problem when trying to communicate. To enable communication between devices

that use electric signals and EM waves, a modem is used. A modem is a
fe

MOdulator/DEModulator that changes electric (digital) signals from a computer

ea

to analog signals for a microwave transceiver or phone line.

modem:
A modem can be used to do the following: MOdulator/DEModulator that
• Boost signals that are too weak to travel from the transceiver to the destina- changes electric (digital)
Pl
Re

tion without losing significant amounts of data along the way. signals from a computer to
analog signals for a
• Use alternate EM frequency bands if multiple communications are taking microwave transceiver or
place on the same transmission medium. phone line.
• Replace a NIC when connecting to a network (when using a phone line to
connect to the network).

Lesson 6: Data Transmission 133

 When remote access is set up on a network, a modem must be configured on the
network that gives users access to network resources. However, if two devices are
within tens of meters of each other, a cable wired in a null modem, or ″dumb″,
configuration can be used in place of a modem.

ial
er
at
y
op
eM
Figure 6-4: Network connection devices and the OSI model.
As you can see in Figure 6-4, each kind of network connection device operates

tC
within specific layers of the OSI reference model. The capabilities and limitations
of each device are directly related to the OSI layers in which the device operates.
Some of these devices are network devices; they connect segments of networks to
each other and are only aware of local network addresses. Devices that operate
above the Data Link layer are thought of as internetwork devices. They connect
No
networks to each other and are aware of addresses outside of the local area
c

network.
In the sections that follow, you will learn the capabilities and limitations of
en

internetworking devices, and you will learn the roles that each device plays in an
Do

internetwork. In practice, the distinctions between these devices are sometimes

unclear. Manufacturers often combine favorable attributes of two or more types of
devices to satisfy the needs of their customers.
se
r
fe
ea
Pl
Re

134
 ial
er
at
Figure 6-5: Repeaters connect two network segments.

y
op
Repeaters
eM
The main purpose of repeaters is to overcome the affects of attenuation. Repeat-

tC
ers work in the Physical layer of the OSI model to provide fast signal
regeneration.
Because of attenuation (loss of signal strength over distance), each type of trans-
mission media used in networks has a maximum suggested length. To extend the
No
network beyond this distance, you can use repeaters. A repeater is typically used
c

as a connection device between two segments on the same network. It can also
be used, however, as a transmission media adapter that connects different media
types. repeater:
A device that connects two
en

network segments to make

Do

them work as one. Repeaters

can extend the length of a
network beyond the physical
limitations of a single cable.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 135

 ial
er
at
y
Figure 6-6: A repeater regenerating a weak signal.

op
eM
Amplifiers
In digital computer networks, repeaters aren’t simply amplifiers; they’re signal

tC
regenerators. If repeaters only amplified a signal, they would also amplify any
noise on the data-transmission medium. Instead, repeaters deal with attenuation
This would be like turning by receiving and evaluating the data (much like any network computer) then
the volume up while immediately re-transmitting a clean copy of the signal on to the next segment of
No
listening to a distant radio
the network cable.
station. The signal gets
c

louder, but so does the

static.
Signal-regenerating Repeaters
en
Do

This type of repeater boosts a signal and produces less attenuation by receiving
the data and removing it from the transmission signal. It then puts the signal back
together and resends it on the next network cable segment. The signal is an exact
replica of the original, including its original strength.
Regenerating a signal takes very little time. However, LANs don’t function well
se
r

if a significant amount of time elapses between signal transmission and reception.

This period of time, called a propagation time delay , shouldn’t exceed a specific
fe

limit. Introducing several repeaters into a network can cause an excessive propa-
ea

gation time delay. For this and other reasons, there is a limit on the number of
repeaters that a network can use. This number is called the repeater count . The
repeater count depends on the type of data-transmission medium and other related
factors. Although the repeater count you can use is limited, you can use repeaters
Pl
Re

to extend the span of a LAN well beyond the normal limitations of the data-
transmission medium. Network segments connected by a repeater must be
identical (for example Ethernet to Ethernet or Token Ring to Token Ring).
Devices that perform the functions of a repeater include Ethernet multiport
repeaters, active hubs, and concentrators.

136
One common type of repeater amplifies the signal and is often referred to as an
amplifier ; another type regenerates the signal and is often referred to as a signal
regenerating repeater .

ial
er
at
y
op
eM
Figure 6-7: Bridges can filter inter-segment traffic.

Bridges
tC
A bridge , like a repeater, extends a network’s transmission distance. However, a
No
bridge is used to connect segments on different networks. It uses a recipient
node’s address to determine which signals should be passed to another network
c

segment (see Figure 6-7).

For example, if the bridge shown in Figure 6-7 receives a data unit from Segment
en

1, it examines the destination node address for that data unit. If the data unit is
Do

destined for a node on Segment 1, the bridge ignores the data unit; the data unit
has already found its way to the correct network segment. If the data unit is des-
tined for a node on another segment, then the bridge forwards the data unit to the
next segment.
In this scheme, traffic local to a particular segment is confined only to that seg-
se
r

ment; however, internetwork traffic is allowed to pass through the bridge. You can
use bridges when you need to restrict traffic across network segments.
fe
ea

Hubs
Depending on how your network is configured, you might need a single point
Pl
Re

where media segments can connect. This single point is called a hub . A hub acts
as a go-between for devices connected to a network; it receives transmission sig-
nals from one device and sends them out to other connected devices. It has
multiple ports to which network devices attach.
There are different kinds of hubs, including active and passive hubs, multiport
repeaters, and switches.

Lesson 6: Data Transmission 137

 Active Hub
Like a repeater, an active hub can regenerate or amplify a signal. Because of this,
and because it connects multiple media segments, the active hub extends cable
length and thereby increases transmission length. Computers attached to an active
hub continue to receive transmissions from all other network computers. In other
words, the hub does not perform any filtering of packets; it broadcasts everything

ial
over every segment.

Passive Hub
A passive hub resembles an active hub only in that it connects multiple media

er
segments together. Aside from that, they differ greatly:
• A passive hub is not a repeater; it does not regenerate or amplify a signal.
• Restrictions placed on segments attached to a passive hub are different from
those placed on segments attached to an active hub.

at
• Additional restrictions can be placed on segments attached to a passive hub.
For example, prohibiting two passive hubs from being consecutively
connected.

y
op
eM
Multiport Repeater
A multiport repeater is similar to an active hub; it receives signals and regener-

tC
ates them. However, a multiport repeater transmits the signals not only to the
intended receiver, but also to every port, regardless of address.
No
c
en
Do
se
r
fe
ea
Pl
Re

138
 ial
er
at
y
op
eM
Figure 6-8: Switches control which segments are connected to each other.
tC
No
Switch
c

Contrary to a multiport repeater, a switch receives a transmission and retransmits

it using only that port through which the intended receiver can be reached. This
en

is very similar to a bridge.

Do

Switches can make efficient use of bandwidth; all of a network’s media segments
can be permanently connected. Each segment is used only for those transmissions
targeted at a computer on that particular segment.
se
r

Multiplexers
fe

Sometimes bandwidth is not used efficiently because a signal is not large enough
ea

to use all the space provided by the transmission media. To avoid this, you can
install a multiplexer, which runs multiple signals on a single transmission
segment.
Pl
Re

Lesson 6: Data Transmission 139

 Multiplexing
In many cases, data-transmission facilities can be compared to postal services.
For example, consider how expensive it would be for people who have never
mailed a letter or packet to hire their own private postal workers, purchase their
own delivery trucks, airplanes, and so forth. Furthermore, consider how those
facilities would be wasted. With the small volume of mail that one customer

ial
could generate, those facilities would be idle most of the time.
As ridiculous as this scenario seems, this type of situation could easily happen
with data-transmission facilities. Often, a main communication line provides a
high-bandwidth (high-volume) connection between two points. Devices on one
end of the line need to communicate with devices on the other end. None of

er
those devices is likely to need the entire bandwidth of the communication line. It
would be expensive and wasteful to have a dedicated line for each set of devices
that needed to communicate.
Multiplexers enable several devices to simultaneously share a high-bandwidth
line. They do this by combining several low-speed channels at one end into a

at
single high-bandwidth channel (multiplexing or muxing), and then splitting (de-
multiplexing or demuxing) the channel at the other end back into low-bandwidth
channels. A multiplexer at either end of the high-bandwidth line performs the

y
multiplexing and de-multiplexing functions. A multiplexer is often referred to as a
mux .

op
eM
To enable devices to share a high-bandwidth line, a multiplexer employs one of
the following schemes:
• Frequency-Division Multiplexing (FDM) —divides a broadband channel’s

tC
wide band of frequencies into several smaller frequency bands of
subchannels.
• Time-Division Multiplexing (TDM) —divides the band on a timesharing
basis.
No
• Statistical time-division multiplexing (STDM or statmux) —designates a time
c

slot for a particular network device according to when the signal accesses
the channel.
en
Do

TASK 6A-1:
Identifying network connection devices
se
r

Objective: Figure out which network connection device is described.

fe

1. Converts digital signals to analog and vice versa.

ea

2. Runs several signals on one media segment.

Pl
Re

3. Connects a device to a media segment.

140
4. Receives signals from a device and transmits them to other devices.

5. Connects segments from different networks.

ial
6. Connects segments from the same network and boosts data signals.

er
CHECK YOUR SKILLS 6-1 Suggested time:
5 minutes
Discussing network connection devices

at
Provide a definition for each of the following devices.

1. What is a media segment?

y
op
eM
2. How does a passive hub differ from an active hub?

3. How does a switch differ from a multiport repeater? tC

No
c
en

4. What is a NIC and why is it used?

Do
se
r

APPLY YOUR KNOWLEDGE 6-1

fe

Suggested time:
ea

5 minutes
Applying networking devices
Decide which networking device is appropriate for each of the following
scenarios. Explain why you chose that particular networking device.
Pl
Re

1. McCloskey Associates needs to provide occasional communication between

segments. Although network traffic is high, most communication is among
nodes on the same segment.

Lesson 6: Data Transmission 141

 2. Fenster Co., whose offices occupy two floors of a small building, needs to
extend the length of its Ethernet cable system to connect two offices. There
are no standard network traffic patterns.

ial
3. Condello Concrete Company has an old Ethernet network. They would like
to use their existing network bandwidth better and make it easier to modify
traffic patterns as needed.

er
4. KK Software Corp. has a room in which new software is tested. The eight
computers in the room run their own network and need a device through
which they can connect to each other.

at
TOPIC 6B

y
op
eM
Network Adapters
Network adapters prepare data for the data-transmission media, send data to and

tC
receive data from the network, and translate data into bytes the computer can
understand. Network adapters can also be called network interface cards (NICs),
LAN cards, and interface cards .
No
c
en
Do
se
r

Figure 6-9: A network adapter.

fe
ea

The network adapter fits into a slot in the computer called a bus , and is attached
to the network cable via a connector. The purpose of the bus is to shuttle data
back and forth from a network adapter to memory and on to the central process-
ing unit (CPU). A bus can be defined as the internal pathway that connects the
Pl
Re

microprocessor to random access memory (RAM). Buses are measured by the

number of bits that can be transferred at one time. The larger the number of bits

142
that are transferred at one time, the faster the bus is. Older buses transmitted 8
bits at a time and were known as 8-bit buses. Newer buses are 16-bit and 32-bit
buses. The type of bus in the computer determines which type of network adapter
must be used to connect to the network. There are several types of buses, as
explained in the following table.

Bus abbreviation Bus architecture Description

ial
It is inappropriate to
- PC Bus The original 8-bit bus of the IBM PC and abbreviate MicroChannel
XT computers. Architecture as MCA. MCA
ISA Industry Standard 16-bit system developed in 1983; a is a trademark of Universal
Architecture majority of computers use this design; less Studios, who legally
expensive than EISA. challenged IBM’s use of the
acronym.

er
EISA Extended Industry Standard 32-bit design; compatible with ISA
Architecture standard.
- MicroChannel Architecture IBM proprietary 32-bit design; a very fast
architecture.
NuBus Macintosh NuBus Apple’s 32-bit design; used only on

at
Macintoshes.
VESA Video Electronics Standards 32-bit design; developed to accelerate
Association video displays.

y
PCI Peripheral Component A more comprehensive 32-bit design; the
Interconnect first to offer Plug And Play setup.

op
PC Card PC Card
eM
Integrated circuits that enable a notebook
computer with one port to interface with
several different devices; about the size of
a credit card; fits into a slot in a mobile

tC
computer; has the ability to ″hot swap,″ or
change the card without restarting the
computer. Originally, this bus was known
as the Personal Computer Memory Card
No
International Association (PCMCIA) bus.
c

However, that name was dropped in favor

of the much simpler PC Card.
en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 143

 ial
er
at
y
op
eM
Figure 6-10: Examples of buses.

tC
No
c

Figure 6-11: A PC Card network adapter.

en
Do

Data moves through a computer’s bus 8-bits, 16-bits, or 32-bits at a time across
separate and parallel conductors. Such a transmission method is called parallel
because you could think of the bits as ″moving side-by-side.″ Network cabling,
however, transmits data in serial fashion, one bit after the other down a single
conductor. It is the job of the transceiver (transmitter/receiver) component of the
se

network adapter to translate the parallel data stream of the bus to the serial data
r

stream the network cable can use. Some network cards require an external trans-
ceiver; others come with an on-board transceiver. Typically, adapters that have an
fe

on-board transceiver also support an external transceiver for backward

ea

compatibility.
Pl
Re

144
 ial
er
Figure 6-12: A network adapter with an on-board transceiver.

Wireless Network Adapters

at
Network adapters can also support wireless networks, which means the network
adapter doesn’t need to be physically connected to a network cable as its data-

y
transmission medium. Wireless network adapters for radio-frequency
transmissions often come with an indoor omnidirectional antenna to function as

op
the data-transmission medium.
eM
tC
Role of the Network Adapter
Network adapters are responsible not only for translating data into a form the
network cable can transmit, but also for broadcasting their address to the network. The MAC in MAC address is
Each network adapter has a unique hardware address, often called a MAC (Media named after the Media
No
Access Control) address , which is used to distinguish it from the rest of the Access Control sub-layer of
c

adapters on the network. Most network adapters have their hardware addresses the Data Link OSI Model
assigned to them at the time of manufacture. layer.
en

An IEEE committee determines MAC addresses. Blocks of addresses are assigned

Do

to each manufacturer. Manufacturers then assign addresses to each card from their
assigned pool of addresses. Each card thus has a unique address. A MAC address
is a 6-byte hexadecimal number (at least for Ethernet cards, other technologies
might vary); for example, 00-A0-8F-CA-3C-25.
The network adapter also communicates with the receiving network adapter to
se
r

determine the following information:

• The maximum size of the groups of data to be sent.
fe

• The intervals between groups of data.

ea

• The amount of data that will be sent before the receiving card issues a
confirmation.
• The wait time for confirmation.
Pl
Re

• The data capacity of each card.

Lesson 6: Data Transmission 145

 • The data-transmission speed.
Finally, network adapters assemble data into frames. Frames are the smallest unit
frames: of data sent over the network. Assembling frames is a function of the Data Link
The smallest unit of data layer of the OSI Model. A typical frame has the following components:
sent over the network.
Component Description

ial
Destination ID The MAC address of the device that is to receive this frame.
Sender’s ID The MAC address of the device that sent this frame.
Control Information about the frame, such as its type and segmentation information
(how the data was divided into frames).
Data The data being sent, or at least part of it.

er
Cyclic Redundancy Check An error detection code typically computed by some sort of mathematical
(CRC) operation on the data (such as adding it up). The CRC is re-calculated at
the receiving station to determine if the frame was corrupted en route. If
so, a re-transmission of the data is requested.

at
Performance

y
Although each network adapter is set to meet certain standards, some network

op
eM
adapters provide features that increase the data-transmission speed. The following
table shows the features that can improve network performance.

Feature Description

tC
Shared adapter memory RAM on the network adapter shared by the adapter and the
computer.
Shared system memory A section of computer memory used by the network adapter to
process data.
No
Bus mastering Temporary control of the bus by the network adapter; data
c

bypasses the CPU and moves directly into system memory.

RAM buffering A buffer created by RAM chips on the network adapter to hold
the data overflow. Data overflow occurs when the network
en

adapter receives more data than it can transmit to the computer

Do

across the bus at one time.

Onboard microprocessor A microprocessor contained on the network adapter that helps
process data without relying on the CPU.
se
r

Network Adapter Configuration

fe
ea

Network adapters, as well as other adapters, hardware, and applications, must

communicate with the CPU in an orderly and established manner. To this end,
network adapters have configuration options that must be set for the network
adapter to function properly. Four of the most common configuration options are
Pl
Re

interrupt (IRQ), base I/O port address, base memory address, and transceiver (for
Ethernet network adapters).

146
Interrupts
Interrupts signal the system processor that the adapter card, system hardware, or
application software needs attention. When signaled, the processor puts its current
task on hold and responds, or services, the interrupt. The Intel family of computer
chips supports hardware, exception, and software interrupts.

ial
Interrupt Description
Hardware Interrupts that come from system hardware, such as the keyboard or the system
clock, or from adapter cards. One such signal, interrupt number 25, causes the
processor to reset and the system to reboot. This is what happens when you
press [Ctrl][Alt][Del].

er
Exception Interrupts used mainly by the processor itself to handle error conditions. An
exception interrupt will be issued, for example, when a process requests the
processor to divide a number by zero.
Software Interrupts sent by an application running on the computer. Software interrupts
generally trigger one of the built-in BIOS routines. These built-in routines can
be replaced by routines available from third-party vendors. The substitute

at
routines would then be called in case of a software interrupt.

y
Adapter cards have a unique hardware interrupt number to distinguish them from
the other adapters in the system. The term IRQ , or interrupt request line , can be

op
eM
used to describe the interrupt number to be used by an adapter. Should two or
more adapters share an interrupt number, the processor will be unable to deter-
mine which adapter sent the interrupt. In such a situation, the processor might
attempt to service either or neither of the adapters, usually resulting in erratic

tC
behavior. (Some new buses support shared interrupts; the adapters used in such
configurations must support the same technique for sharing interrupts.)
In most cases, IRQ3 or IRQ5 can be used for a network adapter. IRQ5 is the
default setting on most computer systems. On Microsoft systems, you can use
No
Microsoft Diagnostic (MSD.EXE or WINMSD.EXE) to determine which IRQs
c

are currently being used.

IRQ Common use in an 80286 and higher computer

en
Do

2 EGA/VGA video adapter

3 COM2, COM4, bus mouse; because these devices aren’t commonly used, this IRQ is
generally available.
4 COM1, COM3
5 LPT2; because this device isn’t commonly used, this IRQ is generally available.
se
r

6 Floppy disk controller

7 LPT1
fe

8 Real-time clock (RTC)

ea

9 Cascade; makes IRQs higher than this one available.

10 Available
11 Available
Pl
Re

12 PS/2 port mouse

13 Math co-processor
14 Hard disk controller
15 Available

Lesson 6: Data Transmission 147

 Input/Output (I/O) Addresses
Once the adapter has signaled the processor that it needs attention (with an inter-
rupt), it then needs a way to communicate with the CPU. This is accomplished
with an I/O address. Several terms are used to describe this feature, such as base
I/O address, I/O address, or I/O port. An I/O address marks the beginning of a
range of memory, usually in the lowest portions of memory, that is used for com-

ial
munication between the CPU and the adapter. Each adapter must have its own
unique, non-overlapping I/O address space. I/O addresses are generally ranges of
memory from 4 to 32 bytes of contiguous memory space. These addresses usually
fall between 0x200 and 0x3FF (hexadecimal). Your computer documentation
should include a list of addresses already in use.

er
Port Typical use Port Typical use
200-20F Game port 300-30F Network adapter
210-21F 310-31F Network adapter
220-22F 320-32F Hard disk controller (PS2 model 30)

at
230-23F Bus mouse 330-33F
240-24F 340-34F
250-25F 350-35F

y
260-26F 360-36F

op
270-27F
eM LPT3 370-37F LPT2
280-28F 380-38F
290-29F 390-39F
2A0-2AF 3A0-3AF

tC
2B0-2BF 3B0-3BF LPT1
2C0-2CF 3C0-3CF EGA/VGA
2D0-2DF 3D0-3DF CGA/MCGA, EGA/VGA color video
No
modes
c

2E0-2EF 3E0-3EF
2F0-2FF COM2 3F0-3FF Floppy disk controller and COM1
en
Do

Base Memory Addresses

The base memory address specifies a location in RAM used to provide a buffer
space (also called a control program execution space) on the adapter. For network
se

adapters, most often the base memory address is used for storing incoming and
r

outgoing data. Memory addresses for these on-board memories are usually
mapped to the range between 640 kilobytes (KB) and 1 megabyte (MB), the
fe

upper memory block region of the CPU. Adapters must have unique base
ea

memory addresses, and must operate in the same mode—either 8-bit or 16-bit.
For example, problems can arise when you use an 8-bit network card with a
16-bit video card. In such a situation, the computer might not be able to boot or
Pl

operate properly due to the mode differences. To correct the problem, you could
Re

either set the video card to operate in 8-bit mode or upgrade the network adapter
to a 16-bit card.

148
Direct Memory Access
Direct memory access (DMA) is a technique used for speeding data transfers
within the computer. In a non-DMA situation, the CPU is involved with transfer-
ring every byte of data between adapters and system memory. While dealing with
these data transfers, the CPU cannot perform other tasks. The DMA controller
works with the CPU and other components in the system to speed data transfer.

ial
The controller relieves the processor from this duty and handles all of the related
issues. The DMA controller also communicates with the adapters to facilitate
transferring data from them directly to system memory. Each adapter must use a
unique DMA channel to communicate with the DMA controller. Failure to do so
will prevent proper operation of the adapters. An 8-bit adapter can use DMA
channels 1, 2, or 3, but channel 2 is generally reserved for the floppy-disk drive

er
controller. A 16-bit adapter can use DMA channels 4 through 7, but channel 4 is
rarely, if ever, used and should be considered reserved.

Configuring Adapters

at
Configuration information can be set by using software, switches, or jumpers
—small connectors that tie two pins together to determine which circuits an

y
adapter will use. On older adapter cards, you physically alter circuitry paths on
the card by selectively opening or closing switches. Most network adapters no

op
eM
longer have to be physically altered to set the switches. Software programs now
allow the network adapter switches to be set through user input in the program.

tC
No
c
en
Do

Figure 6-13: Examples of jumpers and switches for an ISA network adapter.
Each brand of adapter (and in some cases each model from the same manufac-
turer) uses different software for configuration. Generally, these applications use
se
r

some sort of signaling scheme on the computer’s bus to detect and then commu-
nicate with the adapter. Some schemes involve a pulsed electronic signal—almost
fe

like Morse code—sent across the bus, to which the adapter detects and responds.
ea

Such direct manipulation of the bus requires direct hardware access. These types
of programs must then typically be run from DOS and not from within Windows
9x or Windows NT.
Pl
Re

Lesson 6: Data Transmission 149

 Boot PROMs
To support diskless workstations, those without a local hard-disk or floppy-disk
boot PROM: drive, some network adapters support boot PROMs. Because these computers
Chip, used on network don’t have any place to store their operating systems locally, they must retrieve
adapters in diskless them from a server on the network. Computers need a special program to enable
workstations, that stores the them to boot, connect to a server, and download an operating system. Such a pro-

ial
program to boot the gram is typically stored on a special chip on the network adapter.
computer, connect the
network, and download the A boot PROM is a chip that you install (or might come pre-installed) on a net-
operating system. work adapter. The PROM in boot PROM stands for Programmable Read Only
Memory and represents a persistent storage location for data or a program. This
chip stores the program necessary to boot the computer, connect to the network,

er
and download the operating system.
PROM:
Programmable Read Only You must purchase a boot PROM specifically made for your brand and model of
Memory. A persistant storage network adapter. Some newer computers with integrated network adapters come
location for data or a with a boot PROM built in.
program.

at
Drivers

y
A driver is a software program that enables a computer to communicate with a
device. Network drivers are written by the network adapter manufacturers to

op
driver:
A software program that
eM
facilitate communication between their network adapters and the network
enables a computer to redirector on the computer. The job of the redirector is to accept I/O requests for
communicate with a device. files and then send them over the network to the requesting computer. Before a
network adapter can function properly, its corresponding driver must be installed

tC
so that it resides on the computer’s hard disk. Each network operating system
will have its own method for installing drivers. After the driver is installed, typi-
cally there are configuration options, like the ones previously described, that must
be set for the network adapter to function properly. The settings you configure for
No
the driver must match those you set on the network adapter.
c

Many times, manufacturers will change a driver to improve performance. You can
update a driver currently installed on your hard disk with a newer driver. You
should check the manufacturer’s Web site regularly for driver updates. The updat-
en

ing process is typically similar to installing the driver for the first time. If you
Do

remove a device, you should also remove its corresponding driver. This will help
ensure that there will be no conflicts between old drivers and any new drivers
you install at a later time. Again, the removal process is typically similar to the
installation process.
se
r

TASK 6B-1:
fe
ea

Defining network adapter configuration options

Objective: Define the following terms.
Pl
Re

1. Interrupts

2. I/O address

150
3. Base memory address

4. DMA channel

ial
5. Driver

APPLY YOUR KNOWLEDGE 6-2

er
Suggested time:
10 minutes
Identifying adapter and bus types
Examine the various available network adapters and bus types.

at
1. Examine the bus connectors and determine what type of connector it is. Apply Your Knowledge 6–2
is optional.
2. Locate any configuration switches or jumpers.

y
op
3. Examine the bus on the system board and determine the types of slots
available.
eM
4. Determine which of the network adapters could be used with the systems.

Network Adapter Installation tC

The installation of a NIC (Network Interface Card) is similar to that of any inter-
No
face card. You must first know what slot you are going to use for the addition of
c

the network adapter. It is also important to have any necessary tools available for
the install. Part of the Network+ certification is to know what to do in certain
situations. This section will prepare you for situations that involve the installation,
en

configuration or troubleshooting techniques used by technicians when handling

Do

network adapter cards.

The following items can be used as a checklist for things you need to do before
getting inside the PC to install, verify that:
• You have the available bus slot in the machine you wish to add a NIC to.
se
r

• The card is compatible with the protocol, the media, and the network operat-
ing system (NOS).
fe

• You have the available system resources necessary for installation including
ea

open IRQ level, I/O address, and DMA channel.

• The necessary hardware and software has been included. Some items to look
for are drivers, utilities, connectors, terminators or external transceivers.
Pl
Re

Though the physical installation is somewhat easy, you must realize the impor-
tance of doing your research before playing ″the guessing game.″

Lesson 6: Data Transmission 151

 TASK 6B-2:
Installing network cards
Objective: To add a network interface card to your workstation if it
doesn’t already have one, or, if it does have a card installed, to

ial
examine the settings.

1. Does your workstation have a network card installed? If so, continue

with step 5. If not, continue with step 2.

2. Obtain the appropriate type of network card for your system and your

er
network. Turn off your system, open the cover, and prepare to install the
card.

3. Is your network card software or hardware configurable? Install and config-

ure in the appropriate order depending on the card. Using the
appropriate configuration method, configure your card to work with

at
If your network card is your system. This might include setting the IRQ, DMA, I/O ports, and on
hardware configured, you’ll combo cards, the port to use.
need to set the DIP switches

y
or jumpers before installing
it into your system; if your
4. Restart your system. Did Windows detect that you have added hard-
ware? If not, run the Add New Hardware Wizard from Control Panel.

op
network card is software
configured, you’ll need to
eM
Follow the prompts to install the new hardware.
install the card before you
can configure it. 5. Using the Device Manager (Control Panel→System→Device Manager),
check the resource settings for your card. Does the Conflicting Device

tC
List show any conflicts for the card? If so, will the card still work?

6. Return to the Windows desktop.

No
c

Troubleshooting Network Adapters

As a network technician you not only need to implement and maintain the net-
en

work but you also need to troubleshoot. This can be one of the most frustrating
Do

jobs for any network administrator.

Diagnosing the Location of the Problem

se
r

You get a hotline call that a user can’t access the network. Where do you begin
your troubleshooting?
fe

The first thing you need to do is determine where the problem is actually
ea

occurring. Just because a user can’t access the network doesn’t always mean that
the problem is at their workstation. Determine if it is related to a protocol, just a
part of the network, all of the systems on the cable, or the whole network. After
you determine where the problem is, compare the differences between the areas
Pl
Re

experiencing problems and those that aren’t. Try resetting or restarting the
affected devices; this is frequently the solution!

152
If it still isn’t working, try the ″divide and conquer″ method of troubleshooting.
This works particularly well when using a Thin-Ethernet network. Divide the
problem area in half and see if the problem has been cured for half of the area.
Keep dividing the area still experiencing problems until you identify where the
problem is occurring. If you find that the problem is actually the network inter-
face card in the user’s workstation, there are several things you can try.

ial
You can run diagnostic software on the adapter. Sometimes these come with the
card or you might use general-purpose diagnostic software. One test that might be
done is a loopback test. You’ll need a loopback plug to plug into the card if the
card doesn’t have an internal loopback test. Data is sent out and ″loops″ around
back into the card. The ″received″ data is compared to the data that was sent to
verify that it’s the same. If it isn’t, then you’ve got a problem with the network loopback:
Data is sent out and ″loops″

er
card.
around back into the card.
You might also check to see if IRQ, DMA, or I/O port settings have changed. If The ″received″ data is
other equipment was added to the system, the current settings might be conflict- compared to the sent data to
ing with what the NIC is currently set for. verify that it’s the same.
There are internal loopback

at
tests and those that require
an external loopback plug to
APPLY YOUR KNOWLEDGE 6-3 test the card.

y
Suggested time:
Determining why a client can’t access the network 10 minutes

op
eM
In this activity, you need to read through the scenario, then determine what you
would need to do so that the workstation can communicate with the network.

1. You just installed a new NIC in a workstation. This workstation has a

tC
modem on COM1, a mouse on COM2, and a plotter on COM3. There are
no local printers on the workstation. The customer has been unsuccessful in
connecting to the network.
No
c

2. A network user got a second printer and now can no longer access the
en

network.
Do

3. You just added a new system to the network that has a 3C509 combo card
se

installed and you can’t access the network.

r
fe
ea

4. Why should you remove drivers for devices no longer part of your system?
Pl
Re

Lesson 6: Data Transmission 153

 TOPIC 6C
Putting Data on the Cable
In this topic, let’s examine ways that data is put on the data-transmission

ial
medium.

Media Access Control

In network communication systems in which more than one computer has access

er
to the same data transmission media, a standard procedure must be established to
prevent or handle the problem of multiple computers trying to send messages out
on the media at the same time. This procedure is called Media Access Control
(MAC) . If two computers attempt to transmit a signal on the same data-
transmission media simultaneously, the signal from one computer would collide

at
with the signal from the other computer, resulting in both signals being destroyed.
Access methods assure that only one computer at a time can transmit data. They
keep the sending and receiving of data over the network organized. There are

y
four common media-access methods: contention, polling, token passing, and
demand priority.

op
eM
Contention

tC
With the contention media-access method , stations listen to the medium and wait
contention: until no other devices are transmitting before they attempt to transmit a signal. If
A media-access method in another station listening to the medium at the same time attempts to send a mes-
which stations listen until no sage simultaneously, then a collision occurs. When a collision occurs, stations
No
other devices are stop transmitting and wait for a brief period. When the channel is clear again,
c
transmitting before they stations attempt to transmit again.
attempt to transmit a signal.
en
Do
se
r
fe
ea
Pl
Re

Figure 6-14: Contention media-access method.

154
Two types of contention media-access methods are in common use: Carrier
Sense, Multiple Access/Collision Avoidance (CSMA/CA) and Carrier Sense, Mul-
tiple Access/Collision Detection (CSMA/CD).

Contention media-
access method Description AppleTalk uses CSMA/CA
and Ethernet uses CSMA/

ial
CSMA/CA The computer sends a short message that acts as a collision detector. This CD.
signals its intent to transmit data. If no collision occurs with the first signal,
then the computer transmits the actual data.This method increases the
amount of network traffic and slows down performance, because each data
transmission must be preceded with an intent broadcast.
CSMA/CD Each computer on the network checks the data-transmission media for

er
network traffic. If it senses that there is no data being transmitted, it can
send data. Once there is data on the media, no other computer can transmit
until the data has reached its destination and the media is free again.
Collisions are detected and data is re-transmitted if necessary.

at
An advantage to the contention media-access method is that it makes efficient use
of the communication channel in low traffic. However, some disadvantages are:
• It doesn’t guarantee computers access to the data-transmission media. A

y
computer could ″hog″ the channel, blocking the access of other computers.
This access is called probabilistic .

op
•
eM
The flow of data in heavy traffic, because of additional computers on the
network, increases collisions.
• Devices cannot be assigned priorities for faster access.

Polling tC
No
With the polling media-access method , a primary device (also called a controller)
c

controls which secondary device has access to the data-transmission media. The polling:
primary device checks each secondary device in turn to see if it has anything to A media-access method in
transmit. Secondary devices are allowed to transmit for a limited period of time which a primary device
en

before the primary device continues polling. controls which secondary

Do

device has access to the

data-transmissions media.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 155

 ial
er
at
y
op
eM
Figure 6-15: Polling media-access method.
The primary device can give priority (extended access to the data-transmission
media) to certain devices, if necessary. Polling is typically used in hierarchical

tC
networks, in which there is centralized control of network communications.
IBM’s Systems Network Architecture (SNA) is an example of a network that uses
polling. Advantages of the polling media-access method are that computer access
to the data-transmission media is guaranteed, and that there is greater network
No
control due to centralization. A disadvantage of the polling media-access method
c

is that it doesn’t use the communication channel efficiently in low traffic; instead,
it ties up excessive bandwidth while listening for and acknowledging messages.
en
Do

Token Passing
Token passing is similar to polling, except that a master station doesn’t control
token passing: access to the data transmission media. Instead, network computers pass a token ,
A media-access method in a special sequence of bits that grants the station holding it permission to transmit
se

which the station with the data. If a station doesn’t need to transmit, it passes the token and the data on to
r

token is the only one who the next computer on the network, which passes it on to the next computer, and
can transmit. so forth, until eventually the token and the data returns to the transmitting
fe

computer.
ea
Pl
Re

156
 ial
er
at
y
op
eM
Figure 6-16: Token passing media-access method.
The transmitting computer checks for transmission errors by comparing the data
it received with the data it transmitted. It then releases the token to the next

tC
computer. In this manner, every computer has an opportunity to transmit.
Token passing is a deterministic media-access method. That is, each computer is
guaranteed access to the data transmission media. This is ideal for networks in IBM’s Token Ring and FDDI
which timing is critical, as in automated manufacturing. Advantages of the token- use the token passing
No
passing media-access method include: it grants guaranteed computer access to the media-access method.
c

data transmission media; it eliminates collisions; and it is efficient under high

traffic conditions. Disadvantages of the token passing media-access method are: it
provides inefficient use of the data transmission media during low traffic; all
en

devices use complex software requiring intelligence; and device software requires
Do

adjustment when you add or eliminate other devices.

Demand Priority
se

Demand priority is a relatively new access method designed for a specific net-
r

work type called 100VG-AnyLAN. In demand priority, the hub determines when demand priority:
each computer can transmit on the data-transmission media. If a computer wants A new media-access method
fe

to transmit, it signals the hub. Such a signal is called a demand . The hub returns for 100VG-AnyLAN networks.
ea

a signal giving permission to transmit one small portion of the data, called a A hub determines when each
packet . When multiple computers want to transmit, the hub cycles through each computer can transmit.
of the requesting computers in port order. Each computer is allowed to transmit
Pl

one packet per cycle. Computers that haven’t signaled the hub that they want to
Re

transmit are skipped in the cycle and don’t take time in the transmission cycle.

Lesson 6: Data Transmission 157

 TASK 6C-1:
Acting out media-access methods
Objective: Identify which media-access method is being described and
identify any drawbacks of each of the media-access methods

ial
described in each of the scripts.

1.
COMPUTER 01: I’m ready to send. I see the data transmission media
is free, so…

er
COMPUTER 01: Hi, COMPUTER 02! How are you today? I’m done
sending…
COMPUTERS 02 and 03: I’m ready to send. Whoops! Someone else is
sending, so I’ll wait.
[Pause]

at
COMPUTERS 02 and 03: I’m ready to send. Whoops! Someone else is
sending, so I’ll wait.

y
[Pause]
COMPUTER 02: I’m ready to send. I see the data transmission media

op
eM
is free, so...
COMPUTER 02: Hi, COMPUTER 01! I’m doing fine, thanks. I’m done
sending.

tC
No
c

2.
en
Do

MEDIATOR: COMPUTER 01, do you have anything to say?

COMPUTER 01: No.
MEDIATOR: COMPUTER 02, do you have anything to say?
COMPUTER 02: Yes, as a matter of fact. Fourscore and...
se
r

MEDIATOR: Your time is up. You’ll have to finish later. COMPUTER

01, do you have anything to say?
fe

COMPUTER 01: No.

ea

MEDIATOR: COMPUTER 02, do you have anything to say?

COMPUTER 02: Yes. (Where was I?)...seven years ago, our...
MEDIATOR: Time’s up. COMPUTER 01, do you have anything to
Pl
Re

say?

158
 COMPUTER 01: No.

ial
3.
COMPUTER 01: I’ve got the token. (Ahem!) Fourscore and ...
Whoops! My time’s up. Here’s the token, COMPUTER 02.
COMPUTER 02: Thanks, but I don’t need it. Here’s the token, COM-
PUTER 03.

er
COMPUTER 03: Thanks, but I don’t need it either. Here’s the token,
COMPUTER 01.
COMPUTER 01: Thanks. I’ve got it. (Where was I?)...seven years ago,
our... Whoops! Where does the time go? Here’s the token COMPUTER
02.

at
COMPUTER 02: Thanks, but I still don’t need it. Here’s the token,
COMPUTER 03.
COMPUTER 03: Thanks, but I don’t need it either. Here’s the token,

y
COMPUTER 01.

op
eM
TOPIC 6D tC
No
c

Network Topologies
A topology is a description of the layout of a specific region or area. A network
en

topology is a description of the layout of the region or area covered by a network

Do

. The complete physical structure of the cable (or data-transmission media) is

called the physical topology . The way data flows through the network (or data-
transmission media) is called the logical topology . While the physical topology
might be classified as part of the OSI Physical layer, it is constrained and typi-
cally defined by the logical topology of a network. Since the logical topology is
se

part of the Data Link layer, the network topology is typically considered part of
r

the Data Link layer.

fe

There are two types of connections that describe how multiple devices connect to
ea

a single cable or segment of data-transmission media. They are point-to-point and

multi-point. Point-to-point connections provide a direct link between two devices;
for example, a computer connected directly to a printer. Multi-point connections
provide a link between three or more devices on a network. All computer net-
Pl
Re

works, regardless of their topologies, rely on point-to-point and multi-point

connections.

Lesson 6: Data Transmission 159

 Common Topologies
Let’s examine some of the common topologies you might encounter on a
network.

ial
Bus Topologies
A bus topology uses one long cable (backbone) to which network devices are
either directly attached or attached by using short drop cables. Because all com-
puters share this bus, a computer checks for any information that might be
coming down the backbone before sending its message. A message passes the
other computers on the way to its destination. Each computer then checks the

er
address of each message to see if it matches its own. Bus topology is a passive
topology , which means that computers connected to a bus topology only listen
for a signal being transmitted over the data-transmission media. They don’t move
data from one computer to the next. Thus, if one computer goes down, it doesn’t
affect the rest of the network.

at
y
op
eM
tC
No
c
en
Do
se
r

Figure 6-17: Bus network topologies.

In a bus network topology , the backbone must be terminated at both ends to pre-
fe

vent the signal from echoing back when it has reached the end of the bus. A
ea

terminator is a resistor that essentially absorbs signals when they reach the end of
the cable. Thus, a terminator is said to have the effect of making the cable look
infinite to a signal. Typically, with shielded cable installations, such as with
coaxial cable, one of the terminators is connected to ground so as to dissipate
Pl
Re

static electricity that might build up in the shielding of the cable.

The advantages and disadvantages of a bus topology are listed below:

160
Advantages Disadvantages
Uses much less cable than other topologies use. A break in the data transmission medium adversely
affects the entire network.
Individual computers can fail without taking down Difficult to troubleshoot and reconfigure.
the entire network.
Easy to install.

ial
Ring Topologies

er
Ring topologies consist of several computers joined together to form a circle.
Messages move from one computer to the next in one direction only. When a
computer receives a message addressed to it, the message is copied and sent back
with a modification to indicate that it was received. Ring topology is an active
topology , which means that computers connected to a ring topology are respon-
sible for moving data from themselves to the next computer. Thus, if one

at
computer goes down, it affects the rest of the network.

y
op
eM
tC
No
c

Figure 6-18: A ring topology.

en
Do

Some networking schemes use a dual ring topology . In these configurations, data
moves in opposite directions around the two rings. In most cases, one ring is the
primary route over which data flows; the second ring serves as a backup in case
of a break in the first ring. In the event of a primary ring break, the primary and
secondary rings are automatically connected to form one larger ring. A single ring
se
r

break won’t bring down a dual-ring network. Such dual ring configurations are
called dual counter-rotating rings , because when un-joined, data flows in the
opposite direction on the secondary ring as that on the primary ring.
fe
ea

Advantages and disadvantages of ring topologies are as follows:

Advantages Disadvantages
Pl

Almost no loss in signal quality over the network Failure of a single connection can take down the
Re

because signals are transmitted by each computer. entire network (unless the more expensive dual-
ring topology is used).
Easy to install. Difficult to reconfigure.
Easy to troubleshoot.

Lesson 6: Data Transmission 161

 Advantages Disadvantages
Topology is unidirectional and continuous (that is,
there is no end of the network to reflect signals).
Reflected signals aren’t an issue.

ial
Star Topologies
The star topology can be used in either Ethernet or Token Ring networks. In
Ethernet, the star topology is referred to as a star-bus. It uses a central device,
called a hub , with drop cables extending in all directions. Each networked device

er
is connected point-to-point to the hub. Hubs can be active or passive . The major-
ity of hubs are active, meaning they regenerate and retransmit the signal. Some
hubs, such as wiring panels, are passive, meaning they act only as connection
active hub: points and don’t regenerate the signal. All messages in a star topology must go
Hubs that regenerate and through the central hub before reaching their destination. Star topology is a pas-

at
retransmit the signal. sive topology. Computers connected to a star topology aren’t responsible for
moving data from one computer to the next. Thus, if one computer goes down, it
doesn’t affect the rest of the network. In Token Ring, the star topology is referred

y
to as a star-ring. It connects workstations via patch cables. MSAUs are connected
passive hub:
together to create the ring.

op
Hubs that act only as a
connection point and do not
eM
regenerate the signal.

tC
No
c
en
Do
se
r
fe

Figure 6-19: A star topology.

ea

Just because a network connects to a central networking device does not mean
it’s a star topology network. Take for instance 10BaseT Ethernet. Like all
Ethernet networks, 10BaseT Ethernet uses a bus topology. However, with
Pl

10BaseT Ethernet, twisted-pair cables run from computers to hubs giving the
Re

appearance of a star topology. Inside the hub, however, each of the cables is con-
nected forming the bus required for Ethernet. The star appearance is just an
illusion.

162
Advantages and disadvantages of star topologies are as follows:

Advantages Disadvantages
de jure standard:
Inexpensive. Low data rate. A standard that exists
Easy to troubleshoot. No de jure (mandated) standards. through codes, laws,
Easy to reconfigure. Moderately difficult to install. decrees, or other forms of

ial
legislation. (Compare with de
Require more cable than other topologies. facto standard.)

de facto standard:
Mesh Topology A standard that exists

er
In a mesh topology network , every computer is directly connected to every other through popular practice.
computer. This topology is rarely if ever used in the real world, though some
incorrectly believe the Internet is a mesh topology network. In a mesh topology
network with an n number of computers, you must have (n * (n-1))/2 two-way
connections. Thus, to connect four computers (n=4) you must have 6 connections

at
(4 * (3))/2 = 6; with eight computers, you must have 28 connections; and with
100 computers, you must have 495 connections. You can probably see why this
method is not frequently used.

y
op
eM
tC
No
c
en
Do
se
r

Figure 6-20: A mesh topology.

fe

Advantages Disadvantages
ea

Every computer is connected to every other A potentially enormous number of connections for
computer, virtually assuring constant networks of even modest size makes mesh networks an
networking communications. unreasonable venture. The many connections necessary
Pl

would lead to exorbitant costs and exceedingly difficult

Re

management burdens.

Lesson 6: Data Transmission 163

 Hybrid Topologies
A hybrid topology is a combination of two or more different topologies. WANs
sometimes have hybrid topologies because they connect a variety of LAN
topologies. Two common hybrid topologies are the star bus and the star ring. In a
star bus topology, there are several star topology networks connected together via
their hubs in a bus topology. The star ring topology is similar to the star bus

ial
topology. However, instead of having the hubs connected together in a linear bus
topology, they are connected to a main hub in a star topology. The big advantage
of hybrid topologies is that they connect disparate topologies. The disadvantage
of hybrid topologies is that they are potentially complex to establish and
maintain.

er
at
y
op
eM
tC
No
c

Figure 6-21: A hybrid topology.

en
Do
se
r
fe
ea
Pl
Re

164
TASK 6D-1:
Identifying Topologies
Objective: Identify each of the topologies illustrated.

ial
1.

er
at
y
op
eM
2.

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 165

 3.

ial
er
at
y
op
eM
4.

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

5. What type of network topology is used in your classroom? What is used

at work?

166
Choosing a Physical Topology
The following characteristics are important to consider when choosing a physical
topology:
• Relative ease of installation, reconfiguration, and troubleshooting.
• The maximum number of units affected by a data-transmission media failure.

ial
The following table compares the characteristics of the bus, ring, and star topolo-
gies:

Bus topology Ring topology Star topology

Overview Because it uses Moderately easy to Installation is moderately

er
established standards, it install, but you are difficult. The design of the
is relatively easy to limited by the network is simple, but you
install. maximum ring length must install a separate
and the number of data-transmission media
devices. for each computer or
device.

at
Installation Installation typically Installation typically Installation requires more
requires less data- requires more data- cabling than most other
transmission media transmission media topologies.

y
(cable) than other than the bus topology.

op
topologies.
Reconfiguration Reconfiguration is
eM
Reconfiguration Reconfiguration is
difficult, especially when becomes more difficult relatively easy. Moves,
distance or number of as the scale of additions, or changes don’t

tC
taps are at the maximum relocation increases. involve more than the
allowable levels. connection between the
computer or device and the
central computer.
Troubleshooting Troubleshooting is Because each device Troubleshooting is
No
difficult, because all incorporates a repeater relatively easy because all
c

computers or devices on (a device that data in a star topology

the network might have regenerates a signal), goes through a central
to be tested to locate the you can easily find point, where it can be
en

problem. cable faults. The distributed to the

Do

downstream repeater appropriate address.

from a failure identifies
the error when it
doesn’t receive a signal
from an upstream
se

repeater within a
r

specified amount of
time.
fe

Media failure All computers or devices Data-transmission Data-transmission media

ea

affects are affected by a data- media failure on a faults are automatically

transmission media single loop system isolated to the failed
failure; failure of a single affects all devices on segment.
computer doesn’t affect the network. However,
Pl
Re

the rest of the network. if you have several

loops, only the
computers or devices
on the affected loop
will go down.

Lesson 6: Data Transmission 167

 TASK 6D-2:
Identifying advantages and disadvantages of network
topologies
Objective: Identify the topologies described in each step below.

ial
1. Installation typically requires less data-transmission media (cable) than
all other topologies.

er
2. All data goes through a central point, where it is distributed to the
appropriate address.

3. Uses one long cable to which network devices are either directly

at
attached or attached by using short drop cables.

y
4. Several computers or devices are joined together to form a circle.

op
eM
5. Messages move from one computer or device to the next only in one
direction.

6. tC
Messages pass through all other computers on the way to their
No
destinations.
c

7. Uses a central hub with drop cables extending in all directions.

en
Do

Suggested time:
APPLY YOUR KNOWLEDGE 6-4
10 minutes
se

Examining network topologies

r

In this activity, you’ll view the Network Topologies presentation on the Network+
fe

Certification CD-ROM.
ea

1. Access the Digital Video component of the Network+ Certification

CD-ROM.
Pl
Re

2. Click on the Network Cables And Devices link, then the Network Topologies
link.

3. Click Overview to begin the presentation.

4. When finished, exit from the Network+ Certification CD-ROM.

168
TOPIC 6E
Bridges
Repeaters amplify the signals they receive and just send them on; they don’t

ial
check the packet to see where it’s headed. Bridges, on the other hand, do read the
packet to find the MAC address. It can then filter the traffic based on the destina-
tion MAC address.
A bridge of any kind—transparent, source-routing, or source-routing transparent
(SRT)—increases the amount of data that passes through a LAN by determining
which frames have a destination on the local network and which should continue

er
on to other LANs. It keeps those that should stay and passes the rest on, much
like a filter system. The frame is then repeated along the next segment toward its
destination. This process extends the transmission media’s length and device
limitations.

at
Transparent, source-routing, and source-routing transparent bridges are each dis-
cussed in the following sections of this topic.

y
80/20 Rule

op
eM
The 80/20 rule is simply a guideline designed to help create an optimally efficient
bridged network. It suggests that 80 percent of your network traffic should be
local, leaving only 20 percent of it to be internetwork. If traffic is local, two seg-

tC
ments connected by a bridge can be used at the same time once the bridge
discovers the receiver’s location. Because of this, it is more efficient and produc-
tive to have most of your network traffic be local.
You could use a repeater, but when doing so, frames can be transmitted on only
No
one segment at a time because they are simultaneously repeated on the other seg-
c

ment, as shown in Figure 6-22.

en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 169

 ial
er
at
y
op
eM
Figure 6-22: The 80/20 rule in regard to bridges and repeaters.

Transparent Bridge tC
Requiring little or no setup, a transparent bridge is invisible to a transmitting
No
device. The sender does not know that its data might have to travel through one
c

or more bridges to a remote segment. For that reason, this type of bridge was
aptly named ″transparent.″ In regard to forwarding data, a transparent bridge per-
forms typical bridge functions—it checks the target device’s MAC address and
en

forwards the data to the appropriate segment.

Do
se
r
fe
ea
Pl
Re

170
 ial
er
at
y
op
eM
Figure 6-23: A frame passing through a transparent bridge.

every segment on a network except the segment from which the frame came. By
tC
Upon establishing a new transparent bridge, it sends every frame it receives to

using a filtering database , the bridge ultimately learns to filter frames. A filtering
database contains the network locations of end nodes.
No
c

Filtering Database
en

The database contains a few permanent entries, such as broadcast addresses,

Do

which are entered by the vendor. Additional entries are made either manually by
the administrator or dynamically as the bridge watches (learns) network traffic.
Eventually, the database contains an address entry for each network device, as
well as each broadcast and multicast address that will be passed between
segments. All of this information is required because a filtering database uses a
se
r

flat addressing scheme.

During its learning process, a transparent bridge looks at a frame’s sender address
fe

and discovers which segment each network devices lives on, as shown in Figure Transparent bridges are also
ea

6-24. Once it learns where each device is, it forwards only those frames that are referred to as learning
traveling to devices on remote segments. bridges because of the way
they ″learn″ network traffic.
Pl
Re

Lesson 6: Data Transmission 171

 ial
er
at
Figure 6-24: Creating a filtering database.

y
The connection point between a bridge and a media segment is called a bridge

op
Spanning tree protocol is
eM
port . A bridge port exists in one of five modes, called port states . The bridge
discussed later in this topic. protocol (spanning tree) or management entity (SNMP agent) tells the bridge port
which state it should be in. When a bridge port is spanning tree, it moves through
the port states, excluding the Disabled state, which does not participate.

A bridge port is similar to a

network adapter in that it •
tC
The five port states, in order, are Disabled, Blocking, Listening, Learning, and
Forwarding.
Disabled. The port does not learn, and is not included in spanning tree
No
attaches to a media segment. operation forwarding. This state is used in bridge management as a way to
c

manually take the bridge port offline.

• Blocking. The bridge is in standby mode. A bridge is in this state when first
started or if it is a backup in a redundant bridge pair. It processes only
en

bridge protocol frames sent to the bridge multicast address.

Do

• Listening. The port is getting ready to learn and forward (the next two
states). It can be in this state only until the confusion of a topology change
dies down and the network is again ″settled.″ During this time, frames are
not forwarded and learning does not happen so that incorrect information
se

will not be added to the filtering database and bridging loops.

r

• Learning. The port can add information that it gathers to the filtering
database. However, it cannot forward frames yet. Like the Listening state,
fe

there is a time limit at this phase that gives the network time to settle down
ea

before frames are forwarded. This prevents unnecessary frame forwarding

and temporary bridging loops.
• Forwarding. The typical, normal port state. It can both learn and forward
Pl
Re

frames. This is the only state in which forwarding occurs.

172
 ial
er
at
y
op
eM
Figure 6-25: The five transparent bridging port states.

Relay Entity
tC
A bridge will not automatically forward all frames. For example, a bridge cannot
forward a frame addressed to the bridge itself. In order for a frame to be for- Transparent bridging
No
warded, it must have a valid Cyclic Redundancy Check (CRC), be traveling to a happens at the MAC
c

remote segment that is included in the filtering database, and include data from sublayer, so it considers
upper layers (the LLC sublayer and up). anything upwards of itself to
be upper-layer.
A service known as the relay entity performs forwarding and filtering for a
en

bridge, provided that the bridge port is in the Forwarding state. The relay entity
Do

uses the filtering database to figure out whether to forward or drop a frame.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 173

 ial
er
at
y
op
Figure 6-26: The relay entity.
eM
While a frame’s CRC is being verified, a transparent bridge reads and holds onto
the message, a technique called store-and-forward. To verify the CRC, the bridge
checks the frame’s length against the value specified within the frame. If the

tC
CRC checks out, the bridge forwards the message.
Store-and-forward mode has advantages and disadvantages. It stops corrupted
It’s important not to confuse frames from being produced on the network, but creates delays or latency, which
increase proportionally with the size of a frame.
No
this type of store-and-
forward process with that of
c

a store-and-forward network;
a store-and-forward network
is a message switching Bridging Loops
en

process that is used for

A bridging loop occurs when redundant bridges are used to give fault tolerance to
Do

delivering email.
a bridged LAN. A bridging loop is a continuous circular exchange of data packets
between two redundant bridges in which the packets are never transmitted to their
destinations. Having redundant paths in a transparent bridge environment can
cause these potentially fatal loops. A broadcast storm (where new packets are
continuously produced by redundant bridges) that occurs when the spanning tree
se
r

protocol is not in use is also referred to as a bridging loop. The following figures
and scenarios explain how bridging loops occur.
fe
ea
Pl
Re

174
 ial
er
at
y
op
eM
Figure 6-27: How a bridging loop is created.
Redundant bridges 1 and 2 connect segments 1 and 2. As Node 1 sends data to

tC
Node 2, Bridge 1 and Bridge 2 each copy the frame from segment 1 and add an Remember that bridges
entry for Node 1 to their filtering databases. The bridges are not physically con- always use the most current
nected to each other, so one bridge forwards the frame before the other. In Figure information.
6-28, Bridge 1 forwards the frame before Bridge 2 does.
No
c
en
Do
se
r
fe
ea
Pl
Re

Figure 6-28: Bridging loops, part 2.

Lesson 6: Data Transmission 175

 Bridge 2, not knowing that Bridge 1 sent the frame, copies it from segment 2 and
checks its filtering database. When it sees that it already has an entry for Node 1
on segment 1, it thinks the information is outdated and changes it to reflect Node
1 on segment 2. In the meantime, Bridge 1 copies the original frame forwarded
by Bridge 2 to segment 2 and thinks its filtering database is outdated. It changes
the database to reflect that Node 1 lives on segment 2 and forwards the frame to
segment 1.

ial
The same situation occurs on segment 1 and the frame is cycled back to segment
2, and so on between the two segments. The frame is never forwarded to its true
destination, Node 2 on segment 2. This happens to all frames between both seg-
ments; eventually, LAN traffic becomes so heavy that the LAN cannot function.

er
TASK 6E-1:
Discussing transparent bridges

at
1. Explain the 80/20 rule.

y
op
eM
2. How do a transparent bridge’s functions differ from those of a typical
network bridge?

tC
No
3. What service filters and forwards data for a transparent bridge?
c

4. You have two redundant bridges on your LAN for fault tolerance. Data
en

is sent from one node to another but never arrives at its destination.
Do

What might have happened? Why?

se
r
fe
ea
Pl
Re

176
CHECK YOUR SKILLS 6-2 Suggested time:
2 minutes
Identifying bridge port states
Identify the following phrases with the appropriate bridge port state.

ial
1. Standby mode

2. Normal state

er
3. Offline

at
4. Unsettled network

y
5. Gathering, but no forwarding

op
eM
Spanning Tree Protocol

tC
The spanning tree protocol arranges segment routes and redundant bridge routes
so that bridging loops do not occur. It is a bridge hierarchy protocol defined by
IEEE 802.1d that includes a root bridge at the top of the tree, and designated
spanning tree protocol:
Arrangement of segment
No
bridges below it. The spanning tree protocol declares one of a pair of redundant routes and redundant bridge
c

bridges the designated bridge, and the other, the backup. This enables the bridge routes to prevent brigding
to identify routes that contain no bridging loops. It also enables redundant bridges loops.
to communicate with each other so that route failures are detected and bridge
en

routing can be dynamically reconfigured around the faulty bridge or segment.

Do

The root bridge is responsible for letting all bridges in the spanning tree know if
something has changed on the network. It periodically issues configuration mes-
sages to all designated bridges from which they can determine whether they need
to reconfigure their routes. The designated bridges forward the messages.
se
r

Spanning Tree Configuration

fe
ea

The method by which the spanning tree protocol enables bridges to communicate
is a packet called the Bridge Protocol Data Unit (BPDU) . The bridges use
BPDU to dynamically choose a root bridge, designated bridges, and backup
bridges for routes between segments.
Pl
Re

Lesson 6: Data Transmission 177

 A spanning tree network contains only one root bridge, which is always the net-
work bridge with the lowest bridge ID (bridge identification number) . Bridge
IDs can be randomly assigned; however, it is best if you manually assign them. A
bridge ID is an 8-byte number of which you assign the first two bytes and specify
the last six bytes to be the assigned port adapter hardware address, as displayed
in Figure 6-29.

ial
When buying a bridge, it’s a
good idea to look into its
management implementation.

er
Figure 6-29: A bridge ID.
It is a good idea to choose a bridge that is physically closest to the network’s

at
center as the root bridge. (Assign that bridge the lowest two-byte number in its
bridge ID.) This decreases the time it takes to reconfigure the network in the
event of segment or bridge failures.

y
op
eM
Root Bridge Selection
Selecting a root bridge starts with all bridges broadcasting a frame to all of its
ports. The frame contains the following fields: Dest Addr, Srce Addr, DSAP,

tC
SSAP, BPDU, CRC, MAC Header, LLC Header, BPDU, and CRC, as displayed
in the Figure 6-30.
No
c
en
Do
se
r
fe
ea

Figure 6-30: Fields included in the root selection frame.

Pl
Re

The BPDU section of the frame consists of constants, flags, a root ID, port cost, a
bridge ID, and other miscellaneous items.

178
 ial
er
at
Figure 6-31: A breakdown of the frame’s BPDU section.
The following table is a breakdown of the BPDU section’s components.

y
Section Component(s) Size Description

op
Constants Protocol identifier 2 bytes
eM
Version 1 byte
Message type 1 byte

tC
Flags TC 1 byte Topology Change notification that is used to
adjust timers during a reconfiguration.
TCA 1 byte Topology Change Acknowledgment.
Root ID 8 bytes The root bridge’s identification number.
No
Port Cost 4 bytes The cost of the route to the root.
c

Bridge ID 8 bytes The ID of the bridge that is sending the

message.
Misc Port ID 2 bytes The determining factor if two bridges have
en

equal bridge IDs.

Do

Message age 2 bytes The amount of time between configuration

messages.
Max age 2 bytes The time after which a message should be
purged.
Hello time 2 bytes The amount of time since the last
se
r

configuration message from the root.

Forward delay 2 bytes The amount of time between states during a
fe

topology change.
ea

Each bridge then puts its bridge ID into the Bridge ID and Root ID fields of the
BPDU field, indicating itself as the root bridge. Each frame’s LLC Control field
Pl

then instructs each bridge to copy all incoming BPDUs. When the bridge does so,
Re

it checks the Root ID field. If the field contains a bridge ID lower than its own, it
stops broadcasting its BPDUs and forwards the BPDU with the lower-bridge ID.

Lesson 6: Data Transmission 179

 ial
er
at
y
op
eM
Figure 6-32: The root bridge selection process.
In Figure 6-32, Bridge 2 has the lowest bridge ID and becomes the root bridge.
Bridge 1, by receiving BPDUs from Bridge 2, realizes that it is parallel to Bridge
1 and becomes Bridge 2’s backup.

tC
Once the root bridge is selected, it sends BPDUs from all of its ports every
couple of seconds by using spanning tree protocol. (IEEE recommends a two-
second default.)
No
c

Designated Bridge Selection

en

Each bridge copies the BPDUs received through its root port (a bridge port facing
Do

the root bridge). The BPDU sent by the root bridge has a field called Port Cost,
which the root bridge automatically sets to 0. (At this point, you have already
assigned all other bridge ports a port cost value.)
The bridges consecutively receive the root bridge’s BPDU and add their config-
ured port costs to the port cost value sent by the root. They copy each other’s
se
r

BPDUs from the segment and check the port costs. The bridge with the lowest
port cost is the designated bridge; the bridge with the higher port cost is the
backup.
fe
ea
Pl
Re

180
Bridge or Segment Failure
If a bridge does not receive a BPDU from the preceding bridge within a certain
amount of time, it assumes that the bridge is malfunctioning and sends a Topol-
ogy Change Notification (TCN) from its root port. This is the only time a bridge
will send BPDUs from its root port after a root bridge has been selected. The
bridge continuously sends the TCN on the root port until the designated bridge

ial
acknowledges it upstream. Eventually, the root bridge receives the TCN and
sends a Topology Change Acknowledgment (TCA) in the BPDUs it sends every
two seconds.
Upon receiving the BPDU from the root bridge, each bridge deletes all non-
permanent entries from its filtering database and resumes operation in the

er
Blocking state. The root bridge sends a BPDU specifying how long the bridges
should remain in this state. The amount of time is set by the administrator and
must be sufficient for all bridges to receive the BPDU containing the topology
change.
After the specified amount of time has passed, the bridges repeat the designated

at
bridge selection process. The new route around the failed bridge or segment is
automatically configured and the bridges, once again, go through each state
(Blocking to Forwarding) after which they restart forwarding frames.

y
op
Spanning Tree Protocol Guidelines
eM
The following list is a set of guidelines to follow when configuring the spanning
tree protocol:
•
•
•
Have redundant routes to every segment.
Do not make frames travel through more than two bridges.
tC
Do not make frames travel through more than three bridges after a topology
No
change (where the backup bridge becomes the designated bridge).
c

An efficiently configured spanning tree protocol should provide you with fast
topology changes, minimal reconfiguration time for those changes, and minimal
en

latency. For example, BPDUs are forwarded only one level from the root bridge,
Do

so topology changes can happen almost instantly. TCNs from backup bridges
have to travel over only one segment to reach the root bridge, so reconfiguring
after a topology change happens quickly. Also, having no more than three bridges
in a route reduces latency.
se
r

TASK 6E-2:
fe
ea

Discussing the spanning tree protocol

1. How does the spanning tree protocol help prevent bridging loops?
Pl
Re

Lesson 6: Data Transmission 181

 2. During root bridge selection, which network bridge becomes the root
bridge?

3. During designated bridge selection, which bridge becomes the designated

bridge?

ial
4. When configuring the spanning tree protocol, do you want to have
redundant routes to each segment?

er
5. What is the minimum number of bridges frames should travel through?

6. What are some benefits of the spanning tree protocol?

at
y
op
eM
Source-routing Bridge
Source-routing bridges differ from transparent bridges in that they do not use fil-
tering databases. Instead, they use dynamic routing tables, which contain paths to

tC
each device a source-routing bridge communicates with. Source-routing bridges
are typically used in Token Ring networks.
The frame that a device sends out contains routing information in its MAC
header. The source-routing bridge uses this information to forward or drop the
No
frame. Unlike transparent bridges, source-routing bridges copy only frames with
c

source-routing information in their headers. (A transparent bridge copies all

frames from a segment.) Even though frames on a source-routing network contain
routing information, they use a bridging process and not a routing process,
en

because they use physical addresses.

Do
se
r
fe
ea
Pl
Re

182
 ial
er
at
Figure 6-33: Frames using a source-routing bridge.

y
In Figure 6-33, Node 1 transmits data to Node 2. In its MAC header, Node 1

op
eM
includes the route by which the data should travel to Node 2 on segment 3.
Bridges 1 and 2 use that information to forward the data.

Route Discovery
The means by which a source-routing device creates a routing table is called tC
route discovery. Route discovery is a process that establishes a route table based
No
on a hello frame between a sending and receiving device. The sending device
c

transmits the hello frame to the receiving device. As the frame travels to the
receiver, each bridge it passes through adds route information to the frame’s
existing route information, dynamically creating a route table.
en
Do

The table is updated when the receiving device responds to the sending device
with the completed information. The sending device adds it to the routing table.
Route tables are dynamic and are usually stored in RAM. When a device’s power
is turned off, route table entries are cleared from the table. If the bridge is not
turned off, entries stay in the table until the station entry is reset or overwritten
when the table limit is maximized.
se
r

In a Token Ring network, devices minimize traffic by attempting to find a device

locally. A frame is first sent out without source-routing information. If it returns
fe

with the Address Recognized Indicator bit specified, the sending device knows
ea

that the receiving device is on the local ring.

Many source-routing networks, such as IBM, use LLC eXchange Identification
(XID) packets as hello packets, but you can use any hello-oriented protocol that
Pl
Re

creates establishes communication between two systems.

Lesson 6: Data Transmission 183

 TASK 6E-3:
Discussing source-routing bridges
1. What is the main difference between source-routing and transparent
bridges?

ial
2. Can route tables be stored in RAM?

er
3. What is the purpose of a hello packet?

at
Suggested time:
APPLY YOUR KNOWLEDGE 6-5
30 minutes

y
Reviewing the Physical layer topic

op
eM
In this activity, you’ll work through some sample test questions that deal with the
information presented in this (and the previous two) lessons.

1. Load the Skill Assessment component of the Network+ Certification

tC
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ text.

3. Select all items for the Physical Layer topic, uncheck the Test Time check
No
box, and then click OK.
c

4. Work through the sample questions and score the test.

en

5. If you want to review the questions and correct answers, click Review Items.
Do

Use the Explanation and Next buttons to work through the questions and
answers; then use the Scores button to return to your test score window.

6. Exit the test.

7. Start the Network+ test again, this time choosing the Data Link Layer topic
se
r

and uncheck the Test Time check box; then click OK.
fe

8. Review the questions and answers as described in step 5.

ea

9. Exit the test.

Pl
Re

184
 Summary
In this lesson, you’ve examined network connection devices, the role of net-
work adapters in network communication, various access methods for
getting data on the transmission medium, and identified characteristics of

ial
various network topologies. You also examined the use of bridges in
networking.

LESSON 6 REVIEW

er
6A Explain the purpose of network connection devices and describe issues
associated with them.

at
y
6B What is the role of a network adapter?

op
eM
6C What function does an access method provide for data transmissions?

tC
No
c

6D Draw an illustration of each of the following topologies: Bus

en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 185

 Star

ial
er
at
Ring

y
op
eM
tC
No
c
en

6E How does a transparent bridge forward data packets?

Do

What is a bridging loop and how does it occur?

se
r
fe
ea

How do source-routing bridges differ from transparent bridges?

Pl
Re

186
The Network Layer LESSON
7

ial
Overview Data Files:
Routing occurs at the Network layer of the OSI model. Routers and brouters Network+ Certification
are used to link subnetworks (and separate networks) together. Only some of CD-ROM

er
the network protocols in use on networks are routable protocols; you’ll
examine which are and which aren’t. You’ll also take a look at the use of Lesson Time:
default gateways and subnetworks, and the use of unique network IDs. 2 hours
Finally, you’ll examine the differences between static and dynamic routing.

at
Objectives

y
To understand the function of the Network layer of the OSI Model, you will:

op
7A
eM
Describe addressing associated with the Network layer.
You’ll see how addresses are used to route information to the appropriate
network.

7B Explain the purpose of network connection devices and describe

issues associated with them.
tC
In this topic, you’ll examine several internetworking devices and see
No
when they are appropriate. For example, if you’re using a non-routable
c

protocol, you won’t want to use a routing device.

7C Describe the various components associated with routing, including

en

route tables, protocols, and devices.

Do

This topic includes information on routers, routable and non-routable pro-

tocols, default gateways, subnetworks, unique network IDs, and static and
dynamic routing.
se
r
fe
ea
Pl
Re

Lesson 7: The Network Layer 187

 TOPIC 7A
Addresses
Networks, especially large internetworks, might provide more than one path for

ial
moving information from one point in the network to another. The Network lay-
er’s basic purpose is to decide which physical path the information should take to
move from its source to its destination. Determining the path to take is called
routing. The path might span the distance between two computers in the same
office or it might span the globe.
The Network layer of the OSI model might appear similar to the Data Link layer,

er
because these two layers are concerned with moving data to the proper location.
The difference is that the Data Link layer’s concerns and addressing issues were
While the Data Link layer confined to a single segment, or link, in the network, whereas the Network layer’s
deals with issues on a single concerns and addressing issues span networks.
segment of the network, the

at
Network layer deals with The network layer:
issues spanning networks.
• Moves information to the correct address.
• Assembles and disassembles packets.

y
• Addresses and routes data packets.

op
•
eM
Determines the best path for moving data through the network.

tC
Logical Address
The manufacturer usually assigns MAC addresses to devices at the time the
devices are built. The administrator, on the other hand, usually assigns Network
layer addresses. This network address is a logical address, sometimes referred to
No
as the address of the cable. This address is of primary concern to Network layer
c

addressing.
The Network layer address distinguishes one network from another, enabling
en

packets to be routed to their ultimate destination and filtered so that they don’t
Do

get forwarded to inappropriate networks. To be able to route packets, all logical

addresses must be unique. To ensure uniqueness across multiple networks, a com-
bination of logical (network) and device (physical) addresses is used.
se
r
fe
ea
Pl
Re

188
Service Address
Some devices can perform several functions simultaneously. These collective
functions are known as entities. An entity must have a unique address so that it
can transmit data. This identifying address is referred to as a service address and
is a combination of a device’s logical, physical, and service addresses. The ser-
vice address is appended to the logical and physical addresses, enabling an entity

ial
to communicate. A service address is sometimes referred to as a port or socket service address:
address. Address (number) of a port
or socket.
In the event that two network processes from different logical networks might
have the same service address, the rule for unique addressing for communication
between these two processes still applies because their full address (service, logi-

er
cal, and device addresses) is used. To identify common network processes, a well known addresses:
group of service addresses, called well-known addresses, are set aside. Reserved service addresses
for common network
processes.

at
Well-known addresses
include the FTP port 21 and

y
HTTP port 80.

op
eM
tC
No
c
en
Do

Figure 7-1: How addresses work together.

se
r

How Addresses Work Together

fe
ea

• The Logical Network address identifies the source or destination network.

• The Physical Device (MAC) address identifies the source or destination
computer.
Pl

•
Re

The service address identifies the specific process that is running on the
source or destination computer.

Lesson 7: The Network Layer 189

 TASK 7A-1:
Examining Network Layer Addressing
Objective: Identify the term being described.

ial
1. Address of the cable only.

2. Combination of service, logical, device.

er
3. Address of the port or socket.

4. Device processes.

at
y
5. Address of the device only.

op
eM
CHECK YOUR SKILLS 7-1
Suggested time:
2 minutes
Why are bridges faster than routers? tC
No
Based on the structure of addresses, explain why bridges are typically faster
c

than routers.
en
Do

TOPIC 7B
se

Internetwork Connection Devices

r
fe

Just like network connection devices physically join segments of a network

together, internetwork connection devices physically join different networks
ea

together. To do so, they use specific hardware designed to connect independent

networks. By connecting networks, users can have access to resources that might
only be available on another network.
Pl
Re

There are three types of internetwork connection devices: routers, brouters, and
CSU/DSU. Although a network is joined to another, it maintains its own charac-
teristics and unique address.

190
Router
Routers, operating in the Network layer, use network address information to
move data through the best path to its destination. While bridges, hubs, and
repeaters connect segments of the same network, routers connect different
networks. Each network must have its own unique address. To connect different A router determines how data
networks, routers must be aware of network addresses. Because routers can deter- should be routed based on

ial
the network addresses in the
mine network addresses, they can be ″intelligent″ about routing data through a
packets it receives.
complex internetwork involving multiple networks.

Subnet router:
A device that connects

er
The independent networks that make up an internetwork are often referred to as networks and can determine
subnetworks, or subnets. A subnet is an independent network physically con- the best path for data when
nected to and part of an internetwork. multiple paths are available.

It is important for each subnet to have a unique logical address so that the router

at
can move data to the appropriate subnet. A logical address is assigned by an
administrator and can be modified at any time, unlike a physical address, which subnet:
is typically assigned by the hardware manufacturer and cannot be changed. Both An independent network
physically connected to and

y
addresses are unique; they help identify a single computer.
part of an internetwork.

op
eM
tC
No
c
en
Do
se
r
fe
ea

Figure 7-2: Routers function as traffic directors between networks.

Pl
Re

Lesson 7: The Network Layer 191

 A router might be a device designed specifically for routing, or it might be a
computer configured (with network adapters and routing software) to serve as a
router. When a file server acts as a router, the file server is said to have an inter-
nal router; routers that aren’t part of a file server are often called external
routers.

internal router: A router functions much like a bridge; however, there are important differences

ial
A file server that functions as aside from whether you transmit data between segments or networks. A router is
a router. more ″intelligent″ than a bridge and it requires more time to determine a data
path, so its processing speed (the time it takes to forward data blocks, or packets)
is typically slower than a bridge’s. However, a bridge cannot determine the most
efficient data path like a router can. The decision to use one or the other should
external router: be based on network needs and environment.

er
A router that isn’t part of a
file server.

Brouter
Because routers are dependent on Network layer communication protocols, they

at
can connect only networks that use communication protocols with which they are
familiar. If a router discovers that a data transmission is using an unfamiliar pro-
tocol, it simply drops the transmission.

y
Brouters are a hybrid of routers and bridges. Brouters function like routers, relay-
ing data transmissions between networks, but when they encounter a data unit

op
eM
that uses a protocol with which they are unfamiliar, they work like a bridge and
brouter: forward the data to the next segment by using a physical address. Because of this
A communication device, ability, you can use brouters for networks on which there is mixed-protocol traffic
often referred to as an and for networks that use protocols that do not support routing, such as NetBEUI

tC
internetworking device, that
functions as both a bridge
and DECnet LAT.
and a router. An advantage of brouters is that they combine the data-handling capabilities of
routers with the versatility of bridges. A disadvantage is that they are generally
No
more complex and expensive than other internetworking devices.
c

Channel Service Unit/Digital Service Unit

en
Do

Because the installation and upkeep of large quantities of transmission media can
become very costly, you might choose to use a public or private service organiza-
tion for your transmission media needs. By using one of these organizations, you
can connect to their media and save yourself the installation and maintenance of
your own.
se
r

Sometimes public service organizations require their customers to use channel

service units (CSU) and digital service units (DSU) in the interest of safety. The
fe

units are designed to shield network users from electric voltages as well as elec-
trical interference. They also ready data for transmission by adhering to any
ea

CSU/DSU: network rules.

(Channel Service Unit/Data
Service Unit) Combines the Both units are parts of a data communications equipment (DCE) device and are
functions of CSU and DSU collectively referred to as a CSU/DSU. A CSU/DSU operates like a modem; how-
Pl
Re

in one device. The CSU ever, it is a digital-to-digital device rather than digital-to-analog. It readies digital
terminates a digital circuit. signals and guarantees that they have sufficient strength and the proper format to
The DSU terminates a data be transmitted over digital WAN links.
circuit to the Data Terminal
Equipment and converts Typically, a CSU/DSU connects to a router or remote bridge by using a V.35 con-
customer transmission data nection or similar synchronous serial interface. It is also known as an integrated
into a bipolar format.
service unit (ISU).

192
TASK 7B-1:
Applying internetworking devices
Objective: Identify which network device would provide the necessary
services for the following services.

ial
1. Stone Investments needs to provide communication between different
networks. Most of the networks use the same protocols; however, some
use different ones.

er
2. Global Travel needs to connect several networks that use the same
upper-layer protocols. They want to use devices that will provide the
most efficient path among the networks.

at
y
3. Andy’s Supply House needs a high-speed WAN connection but can’t
afford the high cost of installation and maintenance that is associated

op
with owning the communication lines.
eM
TOPIC 7C tC
No
Routers
c

Devices on different networks communicate by using routers. Routers, together

with routing protocols, transfer data packets across internetworks. To do so, the
en

router must know the receiving device’s address and a path to the device. To find
Do

out where the receiving device lives, the router uses the device’s software
address, which distinguishes and describes the device and the network on which
it resides. (A device’s physical address describes only the device.)
A software address can be either logical or logical and physical.
se

•
r

Logical —requires mapping to a physical address before the packet can be

transmitted to the receiver.
fe

• Logical and physical —includes both the network address and a copy of the
ea

physical address.

Packet Routing
Pl
Re

In Figure 7-3, Node A resides on Network 2 and wants to transmit data to Node
B on Network 4. Routers 2 and 3 both use the same Network layer protocol. Net-
works 2 and 4 are Ethernet networks; Network 3 is Token Ring.

Lesson 7: The Network Layer 193

 ial
er
at
y
op
eM
Figure 7-3: Routing packets by using Network layer protocols and software addresses.
Node A’s upper-layer processes use the target software address to identify where

tC
the data is going, and then let Router 2’s protocol take over. The protocol sees
the address (Network4NodeB) and knows that the data is traveling to Node B on
Network 4. The protocol constructs a data packet and appends a Network layer
header of the target and source software addresses (Network4NodeB and
No
Network2NodeA, respectively). Realizing that Node B does not reside on Net-
c

work 2, Router 2 enlists the help of the Data Link layer to forward the packet.
The Data Link layer places the data packet in a MAC frame and adds a destina-
tion hardware address for Router 2 since the packet must pass through Router 2.
en

Upon receiving the packet, Router 2’s Data Link layer verifies the target software
Do

address and CRC, deletes the header and trailer, and then gives the rest to the
Network layer.
Router 2’s protocol looks at the packet’s target network section
(Network4NodeB) and sees that the packet must be delivered to a different net-
se

work (Network 4). It consults its route table to find a suitable route for the packet
r

to take to Network 4. Router 2’s route table indicates that Network 4 is only one
hop away and that Router 3 is directly linked to Network 4. Router 1’s protocol
fe

sends the packet back to the Data Link layer, instructing the layer to transmit the
ea

packet to Router 3.
Router 2’s Data Link layer then creates a Token Ring frame (because Network 3
is a Token Ring network) and sends it to Network 3. From there, Router 3’s pro-
Pl
Re

tocol receives the packet and verifies the network information and software
address. Realizing that the packet’s target node resides on Network 4, Router 3’s
protocol checks the packet’s software address for the target node to find out
exactly where the packet must go. The protocol then attaches instructions to for-
ward the packet to Node B and passes it back to the Data Link layer for delivery.

194
Upon receiving the packet, Node B checks the target software address and sees
that it is the intended recipient. It deletes the Network layer header and forwards
the data to its upper layers.

Route Tables

ial
A route table is a database created by a route discovery protocol and containing
network locations as perceived by a specific router. A router uses its route table to
forward packets to another network or router. The action of forwarding a packet
from one router to the next is called a hop. A router sends a packet to another Each router has a table from
router until the packet reaches the network on which its receiver is located. You its point of view.
can specify the number of hops that packets can take from sender to receiver.

er
The number of hops along a route between two networks constitutes that route’s
cost. However, a cost can also consist of other specifications, such as speed. Typi-
cally, a router maintains the most cost-effective route in its table.

at
y
op
eM
tC
No
Figure 7-4: A sample network with three routers: A, B, and C.
c

In the sample network shown in Figure 7-4, the routing tables for Routers A and
B would contain the following information:
en

Routing Table for Router A

Do

Destination Next Hop Type Cost Intf

140.240.0.0 140.240.20.254 Direct 1 1
192.233.66.0 192.233.66.253 Direct 1 2
136.54.0.0 192.233.66.254 Remote 2 2
se
r

126.0.0.0 192.233.66.254 Remote 3 2

fe
ea

Routing Table for Router B

Destination Next Hop Type Cost Intf
192.233.66.0 192.233.66.254 Direct 1 1
Pl
Re

140.240.0.0 192.233.66.253 Remote 2 1

136.54.0.0 136.54.20.254 Direct 1 2
126.0.0.0 136.54.15.254 Remote 2 2

Lesson 7: The Network Layer 195

 Consider network 140.240.0.0 to be network 1; 192.233.66.0 to be network 2;
136.54.0.0 to be network 3; and 126.0.0.0 to be network 4. Notice that each of
the tables above see the network from a different perspective.

Routing Protocols

ial
A routing protocol provides a router with capabilities, such as discovering paths
and services, failure notification, broadcasting route costs, and maintaining route
tables. The protocol resides in a Network layer packet and has no interaction with
Physical or Data Link layer processes. Its only concern is how the networks it is
servicing are connected.

er
Two algorithms are used to describe how a routing protocol creates and maintains
a route table and the method behind selecting a path to a network: distance vector
and link state.

at
Unroutable Protocols
Sometimes a protocol that cannot freely select a route over which to send packets

y
is referred to as unroutable or nonroutable. It is forced to route them by using
only static paths specified in a simple lookup table. This table cannot be modified

op
NetBIOS and LAT are
examples of protocols that
eM
or updated. A protocol that does not use the Network layer to route packets can
also be labeled unroutable.
were originally designed only
with LANs in mind. If you’re configuring a WAN, you can use routable protocols such as TCP/IP or

tC
IPX/SPX. Protocols such as DLC or NetBEUI, which are nonroutable, couldn’t
be used on a WAN.
Bridging and tunneling (encapsulation) are used in conjunction with unroutable
protocols. The following table lists some of these protocols.
No
c

Protocol Is unroutable because

NetBIOS:
Network Basic Input/Output NetBIOS It does not have a Network layer. It communicates by using a Session layer
System. protocol.
en

NetBEUI It is an extension of NetBIOS that is designed to carry data within only small- or
Do

medium-sized LANs.
SNA All of its routes are pre-defined.
NetBEUI: LAT It does not have a pure Network layer. It was designed to cross only a single
NetBIOS Extended User
network and lacks a header field for the logical network address.
Interface.
se
r
fe

SNA: Multi-purpose Devices

ea

Simple Network Architecture.

Networks often incorporate more than one technology in their structures and need
the capabilities of different connection devices. Two devices that combine func-
tions of different devices are a brouter and a switching hub and router.
Pl
Re

LAT:
Local Area Transport.

196
Brouter
A brouter does the jobs of both a router and a bridge but is contained in one
device unit. It is ideal for a situation in which you need to route and bridge mul-
tiple unroutable protocols. However, it is not cost-efficient to use a brouter if you
need only routing or only bridging capabilities because a brouter costs more than
a single router or single bridge.

ial
Switching Hub and Router
As the name implies, a switching hub and router combines the capabilities of a
switch and a router into one device unit. However, switching hubs and routers are

er
based on relatively new technology so there are not many products from which to
choose and those available can cause network problems.

TASK 7C-1:

at
Discussing unroutable protocols

y
1. How are routes selected?

op
eM
tC
2. What constitutes an unroutable protocol? List some examples of
unroutable protocols.
No
c

3. What type of connection device should you use for routing unroutable
protocols?
en
Do

Route Selection Process

se

Once a router’s table has been created, it’s ready for service. By using informa-
r

tion in its table, a router performs an algorithm to calculate the best possible path
for each piece of data it receives. However, the route selection process can be
fe

static or dynamic.
ea
Pl
Re

Lesson 7: The Network Layer 197

 Static Routing
Static selection means the path has been determined—either by a network device,
which places its own header information on the frame, thereby forcing a specific
path, or by a system administrator through his or her own creation of a router
table—the intermediary routers don’t have the authority to make any routing
decisions. Although static route selection is relatively inexpensive, there is little

ial
or no allowance for router failure or overworked routers. Thus, static route selec-
tion can be management-intensive.

Dynamic Routing

er
Dynamic selection uses the router’s algorithms and cost calculations to perform
the latest routing discovery for every packet. Data stops at each router in its path;
that router is responsible for determining the next best step. Dynamic route selec-
tion is highly flexible and responds well to changes in the routing environment
with little or no management. The main disadvantage is the need for complex and

at
relatively expensive routing devices.

y
Comparing Static and Dynamic Routing

op
eM
In summary, static routing is manually configured, and uses a prescribed route
that might or might not be the most efficient route. This method is more secure
than dynamic routing since the route is predictable.

tC
Dynamic routing is automatically configured (except you must manually enter the
first entry) and uses route metrics to determine the most efficient route. This
method is less secure than static routing due to possibly insecure routing paths
being dynamically selected.
No
Dynamic routing tables require less maintenance than static routing tables. How-
c

ever, static routing tables enable the network administrator to have more control
over the contents of the routing tables.
en
Do

TASK 7C-2:
Examining static and dynamic routing
se
r

1. What does static selection mean?

fe
ea
Pl
Re

198
2. What does dynamic selection mean?

3. Compare static and dynamic routing.

ial
er
Default Gateways and Subnetworks
Every network that is part of an internetwork requires a unique network ID so
that each network can be identified. Routers on the internetwork use these IDs to

at
determine where to send messages. All of the hosts within a network use the
same network ID. A router can be specified as the default gateway when config-
uring TCP/IP so that network hosts can reach external network hosts.

y
Each subnetwork is connected to the larger internetwork via a router (or other

op
eM
routable connection device). In order for the hosts on each subnet to be able to
connect to hosts on another subnet, a default gateway must be specified on each
host. This allows the host to reach a router that can then send the data on to
another router or to its final destination.

tC
default gateway:
Subnetworks are most frequently found on TCP/IP networks. Without a default An IP router that provides
gateway you won’t be able to communicate with other networks. Without an access to a remote network.
appropriate subnet mask (an IP address uses a subnet mask to divide the network
into several smaller subnetworks), you won’t communicate with anyone—not
No
even someone on your own local network.
c

Clients need both a subnet mask and a default gateway specified in their TCP/IP
configuration. The default gateway is the IP address of the router on their local
en

network that allows clients to reach beyond their local network.

Do
se
r
fe
ea
Pl
Re

Figure 7-5: How LANs connect using a default gateway to reach external networks.

Lesson 7: The Network Layer 199

 On an IP network, to configure the default gateway, specify the IP address of the
network device that connects the LAN to the external network.
In Figure 7-5, there are two LANs connected via routers using a Frame Relay
connection. The networks are configured using TCP/IP. The default gateway for
LAN A would be the address on the network side of its router, 192.178.200.1; the
default gateway for LAN B would be 192.178.201.1.

ial
Let’s take a look at how to configure these on your workstation.

TASK 7C-3:

er
Examining default gateway and subnets
1. Open the Network Control Panel on your workstation.

2. Display the TCP/IP properties.

at
3. Display the Gateway tab. What is the current gateway? If there isn’t
one listed, why not?

y
op
eM
4. Display the IP Address tab. Is there a subnet mask? Why or why not?

5. tC
Close the TCP/IP Properties window without making any changes.
No
6. Close the Network Control Panel.
c

Suggested time:
APPLY YOUR KNOWLEDGE 7-1
en
Do

5 minutes
Reviewing the Network layer
In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.

1. Load the Skill Assessment component of the Network+ Certification

se
r

CD-ROM. (Remember to accept the disclaimer.)

fe

2. Start the Network+ test.

ea

3. Select all items for the Network Layer topic, uncheck the Test Time check
box, and then click OK.
Pl
Re

4. Work through the sample questions and score the test.

5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

6. Exit the test.

200
 Summary
In this lesson, you’ve examined Network layer concepts, identified devices
working at this layer of the OSI model, and examined routable and
unroutable protocols.

ial
LESSON 7 REVIEW

er
7A Describe addressing associated with the Network layer of the OSI model.

at
7B Explain the purpose of network connection devices and describe issues
associated with them.

y
op
eM
tC
7C What are some of the unroutable protocols? No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 7: The Network Layer 201

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
The Transport Layer LESSON
8

ial
Overview Data Files:
The Transport layer is the 4th layer of the OSI Model. It is responsible for Network+ Certification
making sure a packet is delivered to its destination. It performs connection CD-ROM

er
services for the Network layer. It also provides communication services for
the layers above this layer. Lesson Time:
45 minutes

Objectives

at
To understand the functions of the Transport layer, you’ll:

y
8A Identify Transport layer addressing issues.

op
eM
One of the addressing issues you’ll examine is the purpose of name
resolution.

8B Describe the Transport layer segmenting process.

tC
In this topic, you’ll see how the Transport layer divides messages into
segments.

8C Define the three Transport layer connection services.

No
This topic covers connectionless and connection oriented services.
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 8: The Transport Layer 203

 TOPIC 8A
Transport Layer Addressing Issues
The Transport layer is responsible for reliable communication between the lower

ial
layers and upper layers. This layer organizes data into segments for the lower
layers and then combines the segments into messages for the upper layers. The
upper layers include service providers, such as file, print, and application servers.
The Transport layer:
• Takes action to correct faulty transmission.

er
• Controls the flow of data.
• Acknowledges successful receipt of data.
• Fragments and reassembles data.

at
Some examples of Transport
protocols include NETBEUI, Addressing Methods
XNS, SPX, AppleTalk

y
Transaction Protocol, The Transport layer is often thought of as the heart of the OSI model because it
Transmission Control provides upper layers with a communication channel. To do this, it uses services

op
Protocol (TCP), User
Datagram Protocol (UDP),
eM
provided by lower layers. Through addressing, successful communication employs
conversations between service requesters and clients. Each process running on a
and Windows NT’s Transport network device is identified by a service address specific to the Transport layer.
Driver Interface. The two addressing methods are:

tC
• Connection ID (connection identifier, or socket, or port).
• Transaction identifier.
No
connection identifier:
Connection Identifier
c

Also known as connection ID

socket or port. An address
that identifies each line of Typically, a service provider can converse with more than one client at a time.
The connection identifier or connection ID is an address that identifies each line
en

conversation to the upper

of conversation to the upper layers. The service provider communicates by this
Do

layers.
number, relying on the Transport layer to communicate each address with the
lower layers.

Transaction Identifier
se
r

The transaction identifier or transaction ID is much like the connection ID; how-
fe

ever, it works with smaller pieces of information rather than a whole

ea

conversation. A transaction consists of a request with a response; a conversation

transaction identifier: consists of one or more transactions. Transaction IDs enable the service provider
Enables the service provider to handle the necessary tasks for a request.
to handle the necessary
Pl
Re

tasks for a request. Similar

to connection ID, but works
on transactions rather than a
whole conversation. Also
known as transaction ID.

204
Address and Name Resolution
Computers are designed to function with the use of alphanumeric characters.
Humans, on the other hand, deal better with names. To bridge the gap between
the complicated network addresses, networking protocols offer a service called
address-to-name resolution. Address-to-name resolution is the process of mapping
an alphanumeric address to an understandable name. This service can be per-

ial
formed on every machine in a network or on a specialized service provider
known as a directory or name server. There are two methods for resolving names: address-to-name
the service-requester initiated method; or service-provider initiated method. resolution:
The process of mapping an
• In the service-requester initiated method, the requester submits a packet ask- alphanumeric address to an
ing for a response from the device that matches the submitted name, address, understandable name.

er
or service type. The matching device then responds with either its name or
address.
• In the service-provider initiated method, all service providers send a broad-
cast packet that includes the packet’s name and address. The network
devices or the specialized directory servers gather these broadcasts. With the

at
name and address information saved in a table, the directory server is able to
resolve names and addresses.

y
WINS

op
eM
The name you assign to a computer when you install Windows NT is the comput-
er’s NetBIOS name. For users, NetBIOS names provide an easy-to-use method to
access the resources of computers on the network. Any NetBIOS-based network

tC
utility can connect to a computer by using the computer’s NetBIOS name. How- WINS:
ever, most TCP/IP commands are designed to use TCP/IP addresses, not names. Windows Internet Name
The Windows Internet Name Service (WINS) automatically maps names for IP Service. Automatically maps
NetBIOS (computer) names
addresses in commands and utilities. (Any computer on a TCP/IP network that for IP addresses in
No
performs NetBIOS name resolution is called a NetBIOS Name Server (NBNS). commands and utilities.
c

WINS is one example of an NBNS.)

There are several reasons why you might implement WINS:
•
en

If you have a Windows NT network that is confined to one network seg- NetBIOS Name Server
Do

ment, TCP/IP broadcasts will resolve NetBIOS names to IP addresses. You (NBNS):
might choose to implement WINS in this situation if you want to reduce Any computer on a TCP/IP
broadcast traffic. WINS uses directed datagrams between the WINS server network that performs
and the other hosts to perform the name resolution. NetBIOS name resolution.
WINS is an example of a
• When you have UNIX hosts on the same network with Windows NT com- NBNS.
se
r

puters, you can enter static address mappings for those hosts’ names on the
WINS server. This enables Windows NT users to use commands like ″ping
fe

unix_host″ to communicate with the UNIX systems.

ea

• WINS can greatly simply the administrative burden of name resolution in an

internetworked environment. Without WINS, network administrators need to
maintain manual name resolution tables in LMHOSTS files on all the routers
in the internetwork.
Pl
Re

The WINS Process

The WINS process works as follows:

Lesson 8: The Transport Layer 205

 1. When a WINS client boots up, it sends its IP address and NetBIOS informa-
tion in a directed datagram to a WINS server. Because the WINS client uses
a directed datagram and not a broadcast, the WINS Server can be on a dif-
The WINS database contains ferent network segment from the client.
multiple entries for each
node, because different 2. The WINS Server enters this information in a dynamic database.
services on each computer 3. When a WINS client needs to connect to another computer, the client will

ial
register separately.
query the WINS Server directly to get the IP address for that computer,
instead of broadcasting name queries.
4. The client receives the requested NetBIOS name to IP address mapping from
the WINS Server and stores the information in its NetBIOS name cache.
Thus, the NetBIOS name for each target computer may need to be resolved

er
to its IP address only once via WINS, unless the client computer’s NetBIOS
name cache entry is purged or expires.

TASK 8A-1:

at
Addressing and the Transport layer

y
Objective: To identify addressing issues in the Transport layer.

op
1.
eM
What are two other names for ″socket″?

tC
2. A transaction consists of what two things?

3. List five service providers.

No
c

4. What two addressing methods are in the Transport layer?

en
Do

5. Define connection ID in your own words.

se
r
fe
ea
Pl
Re

206
TOPIC 8B
The Transport Layer Segmenting Process
The Transport layer is responsible for dividing large upper-layer messages into

ial
easier-to-use segments for the lower layers. Conversely, the Transport layer can
also combine many smaller messages into segments for efficient communication
and minimized overhead. To combine smaller messages into a segment, all mes-
sages must have the same destination address.
Messages are made up of the following components: a logical network address, a
physical device address, a service address, connection identifiers, and data, as in

er
Figure 8-1. The components appear in the order that they are listed from left to
right. When smaller messages are combined, their unique connection identifiers
separate their data, as in Figure 8-2.

at
y
op
eM
tC
No
c
en
Do
se
r

Figure 8-1: Breaking large messages into segments for transport.

fe
ea
Pl
Re

Lesson 8: The Transport Layer 207

 ial
er
at
Figure 8-2: Combining small messages for transport.

y
op
eM
Segment Sequencing
When a message is divided into segments to prepare for departure on a packet-
switched network, it undergoes a synchronization process. During the segment

tC
sequencing process, a number is placed on each segment for proper sequencing of
all pieces when they arrive at their destination. At the destination, the process of
segment synchronization is used to organize and return the segments to their
original order.
No
Two factors can make this task a complicated one. First, the amount of memory
c

allotted to the Transport layer might not be adequate. Also, there is a chance of
losing segments. To help decrease the potential risks, connection services are also
performed in the Transport layer.
en
Do

TASK 8B-1:
Reviewing the segmenting process
se
r

Objective: Identify whether each of these statements are true or false.

fe

1. Segments are divided into messages.

ea

2. Only one message can be sent out at a time.

Pl
Re

208
3. A practical function of the Transport Layer segment process is to help
reduce overhead.

4. Segment sequencing assigns sequential numbers to segments before they

are transmitted. Upon arrival at the destination, the out-of-order seg-

ial
ments are returned to their original order.

TOPIC 8C

er
Transport Layer Connection Services
Like the Network and Data Link layers, the Transport layer also provides differ-
ent types of connection services. The Network layer deals with traffic between

at
intermediate nodes and the Data Link layer handles device-to-device traffic. The
Transport layer is concerned with the conversation between two end node
connections.

y
To guarantee reliable end-to-end communication, an acknowledgment might be

op
eM
needed at the Transport layer even though one occurred at the Network layer. The
Transport layer tries to ensure the reliability of data delivery by detecting and
attempting to correct problems that might have occurred in the lower layers (such
as missing or improperly sequenced packets of data). If the Transport layer can-

tC
not actually correct data-transmission problems, it sends a message to the upper
layers that a problem has occurred; then it can request re-transmission.
The specific connection services implemented in the Transport layer are:
•
No
Unacknowledged connectionless service —does not acknowledge data that
c

has been successfully transferred; layers above must provide their own reli-
ability checks (flow, error, and packet sequence).
• Connection-oriented service —establishes and maintains a virtual connection
en

between two nodes. This service performs flow control, packet sequencing,
Do

and error-recovery functions through acknowledgment and reliable end-to-

end connections.
• Acknowledged connectionless service —combines features of the other two
services. It is similar to unacknowledged connectionless service, except that
it provides a mechanism through which individual packets can be
se
r

acknowledged.
fe
ea

End-to-end Error Control

To eliminate or recover from errors-such as segments arriving with duplicate seg-
ment numbers or segments not arriving at all, the Transport layer can:
Pl
Re

• Make the duplicate number unique.

• Use only virtual circuits instead of packet-switching communication services.
• Limit virtual circuits to one per session.
• Use a time-out process to drop packets that have been circulating too long.

Lesson 8: The Transport Layer 209

 The Transport layer can perform Cyclic Redundancy Checks (CRCs) or
checksums to verify that data and header information has not been corrupted.

End-to-end Flow Control

ial
To control the flow of data, the Transport layer uses acknowledgments and either
a guaranteed rate method or a window flow control method to request retransmis-
sion of segments that were not received. There are two types of acknowledgments
The two types of that the Transport layer uses: Go Back N and Selectively Repeat.
acknowledgements used in
the Transport layer include The Go Back N acknowledgment instructs the sender to re-send parts of the mes-
Go Back N, which instructs sage beginning with the packet immediately following the last successfully

er
the sender to re-send parts received packet. The Selectively Repeat acknowledgment instructs the sender to
of the message beginning resend only the packets that were not received, regardless of their location in the
with the packet immediately
following the last
string.
successfully received packet,
and Selectively Repeat,

at
which instructs the send to
resend only the unreceived
packets.
TASK 8C-1:
Discussing connection services

y
op
1.
eM
Define end-to-end flow control.

tC
2. List the four ways to control errors.
No
c

3. How do Go Back N and Selectively Repeat differ in strategy?

en
Do
se
r

Transport Layer Protocols

fe
ea

Transport layers protocols set up the communication session between computers

and ensure that data is able to move between computers reliably. Examples of
popular Transport layer protocols are listed in the following table.
Pl
Re

Transport Layer
Protocol Description
TCP (Transmission Control Protocol) Part of the TCP/IP protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.
SPX (Sequenced Packet Exchange) Part of Novell’s IPX/SPX protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.

210
Transport Layer
Protocol Description
NWLink Part of the Microsoft version of Novell’s IPX/SPX protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.
NetBEUI (NetBIOS Extended User Interface) NetBIOS (Network Basic Input/Output
System) is responsible for establishing the communication session, and

ial
NetBEUI is responsible for providing the data-transport services.
ATP (AppleTalk Transaction Protocol) Responsible for establishing the
communication session and providing data-transport services in Apple
computers.
NBP (Name Binding Protocol) Responsible for establishing the communication
session and providing data-transport services in Apple computers.

er
Connectionless and Connection-oriented Protocols

at
Let’s examine the use of connectionless and connection-oriented protocols.
Connectionless protocols send datagrams (the term for packets in a connectionless

y
environment) across the network without any knowledge of whether they actually
get to their destination. This is a fast method of sending information since there

op
eM
isn’t any setup or tear-down of connections. A connectionless protocol doesn’t connectionless protocol:
A data communication
use guaranteed delivery service in the Transport layer; upper layers of the OSI
model perform this service. If you use a connectionless protocol and a packet method in which frames are
″blindly″ sent over media
gets lost, the Transport layer has no method for finding it—the protocol relies on

tC
without knowing if the
other layers of the OSI model to find it or have it retransmitted. Another name receiver can receive them.
for this type of connection is best-effort connections .
Examples of connectionless protocols include IP, IPX, and UDP protocols.
No
Connection-oriented protocols require that two (and only two) systems create a
c

connection before any data is transferred. The packets are then sent according to
the rules established by the protocol in use. This includes acknowledging receipt
of the packet and re-sending the packet if there is any error. This method is connection-oriented
en

slower than connectionless protocols, but is very reliable. protocol:

A data communication
Do

Examples of connection oriented protocols include TCP and SPX. Also, modem method in which packets are
connections between two systems would be connection oriented. sent and acknowledged by
the recipient and resent if
there is any error.

TASK 8C-2:
se
r

Connection-oriented or not?
fe
ea

Objective: Identify whether each of the following are connection-oriented

or connectionless protocols.
Pl

1. IP
Re

2. IPX

Lesson 8: The Transport Layer 211

 3. TCP

4. UDP

ial
5. SPX

APPLY YOUR KNOWLEDGE 8-1

er
Suggested time:
5 minutes
Reviewing the Network layer
In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.

at
1. Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)
2. Start the Network+ test.

y
3. Select all items for the Transport Layer topic, uncheck the Test Time check

op
eM
box, and then click OK.
4. Work through the sample questions and score the test.
5. If you want to review the questions and the correct answers, click Review

tC
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.
6. Exit the test.
No
c

Summary
en
Do

In this lesson, you’ve examined Transport layer addressing, name resolution,

segmentation of messages, and protocols at the Transport layer.

LESSON 8 REVIEW
se
r
fe

8A Describe the difference between the connection identifier and transaction

ea

identifier addressing methods.

Pl
Re

212
8B Describe the process of segmenting.

ial
8C Define the three connection services in the Transport layer.

er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re