Scribd
Upload
1K views438 pages

(CN) Computer Networks Text Book (2015 Pattern)

computer networks

Uploaded by

rohitmirge09
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
1K views438 pages

(CN) Computer Networks Text Book (2015 Pattern)

computer networks

Uploaded by

rohitmirge09
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
Computer Networks (Code : 310245) Semester V - Computer Engineering (Savitribai Phule Pune University) 3. S. Katre ME, (Electronics and Telecommunication) Formerly, Assistant Professor Deparimeat of Electronics Engineering \ Vishwakarma Insttate of Technology ([Link].), Pune. ‘Maharashtra, India ‘Computer Networks (Code : 316245) (Seamester V, Computer Engineering, Savitribai Phule Pose University) 1S. Katte. CCopght © by Autor. Al sighs reserved. Nop of this publican aay be reprodvced, copied or stoed ina eer system, distributed or transmitted in any form or by any means, including photocopy, recording, or other electronic or mechanical methods, without the prior written permission of tho publisher. “This book is sold subject tothe condition that it shall not, by the way of trade or otherwise, be Lent, resold, hired out, oF “otherwise circulated without the publisher's prioe writen consent i any form of binding or cover other than which itis published and without a similar condition including this condition being imposed on the subsequent purchaser and ‘without limiting the rights under copyright reserved above, First Printed in Indla + January 2001 First Bdition as per New Syllabus + June 2017 (For Savitribai Phule Pune University) ‘This edition is for sale in India, Bangladesh, Bhutan, Maldives, Nepal, Pakistan, Sri Lanka and designated counties in South-East Asia. Sale and purchase of this book outside ofthese countries is unauthorized by the publisher. Printed at : Image Offset, Dugane Ind. Area Survey No. 28/25, Dhayari, Near Pari Company, Pune—41, Maharashtra State, India. E-mail: rahulshabimage@[Link] ISBN : 978-93-5224-597-0 Published by ‘Tech-Max Publications Head Office: B/S, First floor, Maniratua Complex, Taware Colony, Aranyeshwar Comer, Pune ~ 411009, Maharashtra State, India Ph: 91-20-24225065, 91-20-24217965, Fax 020-24228978. Baul : info@[Link], ‘Website : www [Link] (EID: TP407) (Book Code : PO268A] Syllabus. Savitribai Phule Pune University, Pune Third Year of Computer Engineering (2015 Course) (310245) Computer Networks Examination Scheme InSem(Paper) + 30 Marks End-Sem (Paper): 70 Marks ‘To understand the fundamental concepts of networking standards, protocols and technologies, ‘To Jeam differen techaiques for framing, ezor control, flow control and rooting. To leara ole of protocols at various layers in the protocol stacks. To lear network programming, ‘To develop an vaderstanding of modern network architectures from a design and performance perspective, Course Outeomes : ‘On cornpletion of the course, student willbe able to : Analyze the requirements for a given organizational structwe to select the most appropriate networking, architecture, topologies, transmission mediunas, and technologies. ‘Demonstrate desiga issues, flow control and error control. Analyze data flow between TCP/IP model using Application, Transport and Network Layer Protocols, strate applications of Computer Network capabilites, selection and usage for various sectors of user community. ‘strate Client-Server architectures and prototypes by the means of correct standards and technology. Demonstrate different routing and switching algorithms. Router, Brouter and Access Point, Manchester and Differential Manchester Encodings ; IEEB802.11: Frequency ‘Hopping (FHSS) and Direct Sequence (DSSS). (Refer chapter 1) Logical Link Control : Design Issues : Services to Network Layer, Framing, Error Control and Flow Coatrol. Error Control: Parity Bits, Hamming Codes (L1/12-bits) and CRC. Flow Control Protocols : Unrestricted Simplex, Stop and Wait, Sliding ‘Window Protocol, WAN Connectivity : PPP and HDLC. (Refer chapter 2) ‘Medium Access Control : ‘Channel allocation : Static and Dynamic, Multiple Access Protocols : Pure and Slotted ALOHA, CSMA, WDMA, IEEE 802.3 Standards and Frame Formats, CSMA/CD, Binary Exponential Back -off algorithm, Fast Ethemet, (Gigabit Ethernet, IEEE 802. 1afbyg/n and IEEE 802.15 and [BBE 802.16 Standards, Prame formats, CSMA/CA. (Refer chapler 3) Unie | Network Layer : ‘Switching techniques, IP Protocol, IPv4 and IPv6 addressing schemes, Subnetting, NAT, CIDR, ICMP, Routing. Protocols: Distance Vector, Link State, Path Vector, Routing in Internet: RIP ,OSPF, BGP, Congestion control and (QoS, MPLS, Mobile IP, Routing in MANET - AODV, DSR. (Refer chapters 4 and 5) ‘Transport Layer : Services, Berkley Sockets, Addressing, Connection establishment, Connection release, Flow control and buffering, Multiplexing, TCP, TCP Timer management, TCP Congestion Control, Real Time Transport protocol (RTP), (6 Domain Name System (DNS), Hyper Text Transfer Protocol (HTTP), Email : SMTP, MIME, POP3, Webmail, FIP, TELNET, Dynamic Host Control Protocol (DHCP), Simple Network Management Protocol (SNMP). (Refer chapter 7) goo Wh compuwowons srry te toa an Chapter 1: Physical Layer 14110187 ‘Sylebus : Introduction of LAN ; MAN ; WAN ; PAN, Acrhoc [Notwork, Network Archicchures: Cont Sonver ; Peet To Peer, Distibuted and SDN, OSI Model, TCP/IP Modo, Topologies: Star ‘and Herarchical; Design issues for Layers, Tranemssion Maclume: CATS, 52, 6, OFC and Radio Spectrum, Notwork Daviess : Bdge, Switch, Router, brouier and Access Point, (Manchester and Diefental Manchester Encodings ; EEES02.17 : Si 186 127 “128 ~ “429 1.4. Sorvce Provided by the Network fr 14 18 18 in 18 19 1310 1.10.1. Whan to use Peor to Poor Networks 2... 1.10.2 Advantages of Porto Par Networks 1.103 Disadvantages of Peer to Poor Networks... 17 (Glent/ Servar Network (Santr Based Network) AAT Communication in ont Server wit wna 14 111s: iis a7 aT 119 120 Functions of Dierent Layers. seman PB [Exchange of (formation using the OSI Model?33. 1.20.10 Matis o 81 Rare Medel nn 120419 Demers o 08 Mot nan. ‘Computer Networks (SPU) Table of Contents, 1.21. The TCPAP Reference Model. 140 4.21.1 nfroduciion to TOPHP... 140. 1.21.2 Overview of TOPAP Arcitectur, 140. 1.21.3 Daseription of TOPAP Model tt 1.24.4 Comparison of OSl and TOPAP Models... 1-42 1.22 Transmission Media 129 1.22.1 Classiication of Tanséission Media... 130 4.22.2 Comparison of Wired and Wireless Medta... 1-43 1.901 PASSHVG HUD nnn nncrnnsenvnsnn 181 REPOS rnrnninnnenennn 131A ASB HUD aarcveninnn 1.223 Types of Wied Media 1.23 Twisted Pair Cables.. 1.23.1 Types of Twisted Pair Cablos. 1232 Categories (Cal) of UTP. 1.293 Category 9 and Category 5 (Cat 9 and Cat5) UTP CAI ae 1.28.4 Catagory 6 (Cal 6) UTP re nrennennntnn P48 1.923 Compatison of Tranaperent and Souce Routing 1.285 Category 7 (Gat 7) Shielded Screen Twisted Pair B® veneer 5 Remote E1396 enn 132 smn 146 124 125 128.1 Light Souroas fot FID 11.282 Princpte of Light Propagation ina Fiber. 128 1.26.1. Comparison of Sip Index end Graded Indox 1.26. Comparison of Optical Fiber wth Coal and Twisted Pal C888 nn +54 1127 Advantages and Disacvantages of ber Opal Fiber 1-64 427A Advantages of Ocal Fiber... 1272 Oisadvantgos of Opa Fibre. 1273 opeaions 1.41 tnroduction fo Spread Speci. 128 Unguided (Wireless) Medi... 141.4 How's the $8 Signal Difront fom the Nocal 4128.4." Ungulded Meda: Wires, Sonal? 1.40.2 Types Stafons in ESS nen f Networks 1.42 —_Appteations of Spread Spectrum ModUltOR secs 1-78 1.48 Clits be Spd pcm Mean ‘Techniques.. 178 11.46 Frequoncy Hop Spread Spectum (FH-SS) Signal... 1-82 1.46.1 Types of Frequency Hopping 1.47 Siow Frequency Hopping... 147. Operant FHASK Tenant 1472 FHIMESK Rocener. 1.49 Fast Frequency Hopping... 1.49 Advantages and Disadvantages of DS-SS and FHSS. SYSTEM ann 1494 (Snaniiemumaniiinee sve 185, 1492 ‘coment 1502 Conpateon D868 and FHSS Stone 1.95 + Reve Quest ON nen mnnnenmnmennneeen 86 Syllabus : Design lssuoe: Sondoes to Network Layer, Framing, €tror Conta and Flow Cont or Coir: Paty Bits, Hamming Codes (11/12) and GA. Flow Cont! Protocols: Unvestited Simplex, Stop and Wak, Siding Window Protocol, WAN 2.44 Posiion of Data Unk Layer 22 ala Link Layer Design Issues 24 25 26 a7 28 29 210 an axe 213 24 28 Table of Contents 249 Staring and Ending Character win Charactor 244 “ 245 Staring and Ending Fags, wth Bil Stuffing... 2-5 2.4.6 Physical Layer Coding Violations. Error Conta. 254 Error Detection and Corecton.. 264 262 263 264 285 286 267 208 269 2810 2811 Elementary Data Link Protos 2.8.1 An Unvestictd Simpiex ProtOCD nme 2a 283 High Lovel Data Link Conta (HDLC) Protce.. 2.11.1 Frame Stuetur in HDLC. 2412 Frame Typesin HDLC. 2.113 Transparency in HOLC.. a \Whye CRC n Daa Link rotocasin Tailor and nt Hoader 2... se 383 2.158 Network Control Protocal (NCP)... 2.18.9 Mustink PPP. ee 2.15.10 Diltorence between SLIP and PPP. 2.16 Solved Examples... 310 318 319 319 349 Syllabus : Channel allocation: State end Dynamic, Mutiple ‘Access Protocels: Pure and Slotted ALOHA, CSMA, WOMA, IEEE. 802.3, Standards and Frame Formals, CSMACD, Binary Exponent! Backolf algositim, Fast Ethomel, Gigabit Ethernet, IEEE 002.11a/bfpn and IEEE 802.16 and IEEE 602.16 Standards, Frame formats, CSMACA 94 inoduetion.. ve B14 MAC and LLC SubbaY618 nnnen 3.2 The Channol location Probe nesnmrenmesennnan $2 32.4 Stale Channel Allocation in LANG and MANS.. 2 322 Dynamic Channel Alocation aa 312 219 ate 248 33 3.18.1 Swichad EDOM EL oor 9152 Full Duplex EROME enon 9.16 Fast Ethomet tea ee 3.17.4 Physical Layor Implomentaton.. 3472 Ten Gigabit theme. 38.18 Token Bus = IEEE 802.4en cove 3.19 Token Ring System NEE 02.5) 9.18.1. Comparison of Acoass Control Methods... 3-90 3.182 Comparison o 802, 802.4 ane 002.5 IEEE 330 3.193 37 97.2 Wavelength Division Multiple Access Protocols (wows)... 313, 3.223 Wireless LAN - 802.11 (AcitectN9} 323 Technical Issues. 328 320 Sees ‘Components in «Typical IEEE 802.11 Network. 3261 Basic Serios Sat (BSS)... 9.26.2 Types of Staion ESS. nnn 940 Services Ottored by a Typical IEEE 802.11 Notworknn 240 274 AP Sonioas.. 3272 STA Semiees. IEEE 902.11 Stondard, 204 \Wiroloss LAN Protocols, 88 40 eons (Bluetooth ALCHECHU) nnn nermenne 283 ‘Scatleets (Bluetooth Arcitecture) and KSt08 ane 289 ink Manage Proto (LMP) nr SA 3.40.1 Power Management. panne 9A 2423 Comparison of Buotooth and WLAN TEBE 902.116 nen ; 3.43 Applications of Bustoath 3.44 IEEE 802.18 (WMANS) 344.1 The 802.16 Protcal Stack 3442 802.16 Frame Format... 3.44. Diflerence Between IEEE 802.11 ang 41 to 4a7 Chapter 4: Network ‘Syllabus + Switching techniques, Routing Protoedts: Distance Vector, Link State, Pats Vector, Congestion contol a ‘Store and Forwand Packet SWHCHiNg annen 42 ‘Services Provided tothe Transport Layer... 43 423 424 ‘Compasson of Vital Crcult and Datagram 44 45 az aso 4108 — Cxcult Smiches Technology in Telephone Networts . sn AAS 4.11, Telegraph Networks and Massage Switching. 415 AANA A096 enn 48 4112 Disadvantages... 415 416 416 sz “a7 4.3 A138 Cra etched Techy MWA 628 4.196 Advantagos of Virtual Circuit Packet 4138 Comparison of Datagram and Vital Cults 4-29 4.14 Muticast Routg.. 4.15 Routing Algorithms, 4.154 Desied Properties of a Routing Algor... 428 4.182. Types of Rating NDF wen mnnnnennr #24 4.183. Optimality PRC vnernnnnnn 425 AAG) Static (Non adaptive) AIH nonnnnntenn 425 4.164 * Shorist Pab Rowing. 4162 Oikotate Alport... es 4.17 Dynamic Routng Algothrs. 4.17.4 Distance Vector Routing Algor 4.17.2 Countto innit Problem. 4173 Unk Sate Routing 4174 Compson of Unk Stale Routing and Distance Veto PO en 418 Hisrarctical Routing. 419 na and ntrdomain Routing. 420 Intometworking. 4201 Why kuemotworing 4202, The Problems i ilometet nnn AST 4203 Dealing wih Inept 190. AB 421 Frogmentaton 421.1. Statogy- fr Fragmontation (Transparent nS) 4.23.4 Principle of Congsstion CONG nn nnnnnen AAS Syinbus = IP Protocol, Iho and JPVE adessing scheres, ‘Subnoting, NAT, CIDR, KOMP, Rowing in nemet: RIP, OSPF, GP, MPLS, Mot I os 51 intoducion... 514 Why Aas 7. 512 Logical Adresses ( Addresces) 52 ARP (Arse Reeciton Proc. 50.1 Mapping oP Ades ita MAC tes. 87, 59 5.10 581 Fragmentation 552 Auhentcafon ond Pay... 553 Migrating iPv6 (Company to Pes). 554 Comparticn betwoon v4 and Pv6. ‘Mapping Phytal Adirosa to Logical AdGEE8S uuu 520 8.71 The Reverss Address Resolution (RARP} 58.3. _Diforont Types of Messages in ICMP... 28 Unicast Routing Protas FIP (Peutng Ilorin ProtCCD ocnne S101 AIP Updating Agortun..nnonn 510.2 ilataing te Routing Tab. 5103 Updating te Routing Table sen 5D 529 50 50 530 Path Vector RUN nen ST Path Voctor Message. sia 54 5.141 MPLS Hear oa) 5142 How dose MPLS Wot Pec 4D een 0 550 B15, 5.182 5153 8.416 Table Divan Routing Protocols Proec. a7 g Computer Networks (SPPU) Table of Contents $172 Adee on Demand Distance Vector Routing Protocal (AODY).. © Revo Cube nnn 558 5.18 Universty Questions and Anewors.... 558 69.2 Ports and SoCkBES even 610 Features of TP. 611 ToPSenicas. Syllabus + Services, Berkey Sockets, Adcrssing, Connecson |] 6.12 TOP Festus... ‘establishment, Connection release, Flow control and bufering, 6.13 The TCP Protocal 0. Mutiploriog, TOP, TOP Timer manayomen!, TCP Congsston 6.181 TOP Segment Control, Real Time Transport protocat (RTP), Stream Contr Transmission Protocol (SCTP}. Qualty of Service (00S), Dilereiate cores, TCP and UDP tor Wireless eu et 8.16:1. TOP Connection Esatishent 2 161142 Connection Terminabon Proloca! 615 616 825 Comocton Oreted Versus Connoctonless 6.161 Nagi’ Algoit.. S08 en see 4 $26 elma Taped Layer Vere at | \ 63 6172 Inieomet Congestion Conte got ..n650 64 65 618 65.4 Conecontss arate Server a10 68 8.7 Tho otemet Transport Protools (TCP and UDP)... 6412 68 User Datagram Protec! (UDP... en 18 6.8.1 Well Krown Ports for UDP. e1 [RTP (Raal Timo Protocol 6.202. ATCP [ ATP Contra Protoon e2t 622 6.221 Flow Cheractostcs. 6.222 Techniquas for Achioving G00d 008 naan 648 6.229 Trafic SHIN ene arnennynenannnnnn 47 6.22.4 Leaky Bucket Algor... 6.225 Token Bucket Again. 6228 Combination of Token Bucket and Leaky Buck. sn OAD 6227 lasouive Reservason. 623 Scheduling and Potcing. 6.23.4 Unk Scheduling Disc enn OBO 24 POLY nan nas A 625 Integrated Services and Ditronatod SerH06 enn 59 6281 IEOV i esenvnnanmnnennnnn 68 8252. Claseos of Serce. 6.25.3 Diforantiated Sevc28 (OMBGN) een 8S 7410 7-50 13 TA 75 78 7 Appation LAY nnn 744 Chont Server Model... TAZ Mressing 794 mall Arctiocre and S0°4088 wore TAO ‘Sylabus = Domain Name System (ONS), Hypar Text Transfer Protooot (HFTP), Emait SMTP, MME, POPS, Webmaa, FTP, ‘TELNET, Dynamo Host ContolProtocl (HCP), Simple Network ‘Managemont Protoca (SNMP). mH Message Azcass Agant: POP and IMAP... TAA POPS evenness 7at2 MAPS, Computer Networks 10. 72 7495 7.444 Comparison of FTP and TETP 7.18 World Wide Wab (WWW) 7.48.1 Web trom ho Users Sid... 7.152 Web rom the Serves So... 7.163 WWW Archtectura Ot 7.184 Sever. 7462 Uniform Resource Locator (UAL). 7.183 Cockles : User Servet I8r880N ene P28 rar 728 728 7.72 HTM. Hypo Mat Language)... 728 7173 rata 718 7.484 7482 The Web and HTTP 7.18.1 Non-potsisiont and Persistent Connection... 7-28, 7184 7185 7186 7187 7188 7189 718, 720 721 722 723 Host Configuration : DHCP.. 7288 724 Networks 1 7.23.10 DHCP Massage Formal... 71 on TD Config AOD nn 742 7244 Stalic Address Alivcaton... 7242 Dynamic Address Alocation 7.243 Transition States. 7243.1 Aditoss Acquistion Sie... 7243.2 Cary Lease Termination neem none 748 7243.3 Losie Ranowal States... 7244 DHCP Arctitetue.... - 7245 7247 BOOTP Relay Agents. 7.28 Universty Quastions and Answers... Table of Contents 7.25 knoe Network Management Protocol (SNMP)... 7-47 7254 _ vn TAT ‘7.252 Managers and Agents. 74 7253. Management Components. 148 7254 Stud of Management fformaton (S)..7-48 7285 Management information Base (MIB). 7258. Simple Network Management Protocol (SNMP) + Review Question ans gua Physical Layer Introduction of LAN ; MAN ; WAN ; PAN, Ad-hoc Network, Network Architectures : Client-Server ; Peer To Peer, Distributed and SDN, OSI Model, TCP/IP Modal, Topologies : Star and Hierarchical; Design ‘ssues for Layers, Transmission Mediums: CATS, 5e, 6, OFC end Radio Spectrum, Network Devic Bridge, Switch, Router, brouter and Access Point, Manchester and Differential Manchester Encodings : (EEE802.11 : Frequency Hopping (FHSS) and Direct Sequence Introduction : Network : Protot Network is a broad term similar to “system”. Network is @ communication system which supports many users. In relation with the computers we can say that a “computer network” is 2 system which allows conununication among the computers connected in ‘the network. ‘There are various ways of intercoonecting the computers. col : For successful communication to occur, it is not enough for the “sender” to simply transmit the message and “assume” that the “receiver” will receive it properly. ‘There are certain rules that must be followed to ‘ensure proper communication. A set of such rules is known as a “protocol” of the data communication system, ‘Many different protocols are used in the modern data ‘communication system, ‘The interconnection of one station to many stations is called as networking. A network is any intercomnection of two or more ‘tations that wish to communicate, ‘Node : Each station in a communication network is called as a node. The nodes are connected in different way to esch other to form a network. ‘One of such networks is shown in Fig. 1 Many other forms of intereconections are possible, ‘The most familiar network is the telephone system. It 4s the largest and most sophisticated network ofall. (08ss). eh (G19 Fig. LL +A simple communication network 1.1.1, Introduietion to Computer Networks : + During 20° century the most important technology hhas been the information gathering, its processing and distribution. + The computers and communication fields have been merged together aad their merger has had a deep impact on the taanner in which conaputer systems are ‘organized. + The old model in which a single computer used to serve all the computational needs of sn organization hhas been replaced by a new one in which a large ‘number of separate but interconnected computers do the job. * Such systems are called as computer networks. + Two [Link] said to be interconnected if they interchange iaformation. The connection between the ‘separate computers cau be done via a copper wire, fiber optics, microwaves or communication satelite, Distributed system : * A system with one control unit (master computer) and many slaves, or a large computer with remote Printers and terminals is not called a network, itis called « Distributed System. © In distributed system the existence of multiple autonomous computers is not visible to the user. * With a computer network, the- user has to consciously Jog onto a machine, submit jobs remotely, move fies around etc. in short handle all the network management personally, + With a distributed system nothing of this needs to be done explicitly, it all happens automatically because he system takes care of it without the users knowledge. + Basically a distributed system is. a software system baile on top of a netwoek. The software gives ita high degree of cobesiveness, (homoginity) and transparency tothe syste, 1,2 Network Topology Types : + The word physical network topology is used to explain the manner in which 2 network is physically connected. + Devices or nodes in a network get connected to each ‘ther via communication Hinks and all these links are ‘elated to each other in one way or the other. ‘+The geometric representation of such a relationship of links and nodes is known as the topology of that network, + The five basic network topologies are as shown in Fig. 121. ‘+ These topologies can be classified into two types : 1. Peer to peer 2. Primary secondary . Peer to peer is the relationshig where the devices share the link equally. The examples are ring and mesh topologies, © In Primary ~ secondary relatioilihip, one device ‘controls and the other devioes have to transmit through it, or example star and tree topology. 1.2.1 Bus Topology : . ‘The bus topology és usually used when a network under consideration is small, simple or temporary as, shown in Fig. 1.2.2. Node! Nods2 Malls Made (19 Fig 1.22: Bus topology, © On a typical bus network a simple cable is used without additional electronics to amplify the signal Networks 12 Physical Layer ‘or pass it along from computer to computer. “Therefore the bus isa passive topology. © When one computer sends a signal om the cable; all the computers on the network receive the information. However only the one with the address ‘that matches with the destination address stored ia the message accepts the information while all te ‘others reject the message. ©The speed of the bus topology is slow because only ‘one computer can send a message at a time. A computer must wait until the bus is free before it can transmit. ‘+The bus topology requires @ proper termination at both the ends of the cable in order to avoid reflections, + Since the bus is a passive topology, the electrical signal from a transmitting computer is free to travel ‘over the entire length of the eable, ‘+ Without termination when the signal reaches the end of the cable, it returas back and travels back on the cable. ‘©The transmitted waves and reflected waves, if they are in phase add and if they are out of phase cancel. ‘© ‘Thus addition and cancellation of wave results in a standing wave, © The standing waves can distort the normal signals ‘which are travelling along the cable, This can be avoided by terminating the bus on both ends in 50.0 Toad innpedance. + The terminators absorb the electrical energy and avoid reflections. ‘Characteristics of the bus topology : Following are some of the important characteristics of the ‘bus topology : 1. This is a multipoint configuation. There are more than two devices connected to the medium and they are capable of transiting on the medium. Hence the Medium Access Coatrol (MAC) is essential for the bus topology. 2. The signal strength of the tansmitted signal should be adequately high so as to meet the minimum signal strength requirements ofthe receiver. Networks 13 3, Adequate Signal to Noise Ratio (SNR) should be ‘maintained for better quality reception, 4, The signal should not be too strong, This is necessary {o avoid the overloading of transmitter and hence the possibility of signal distortion. 5. Thisis called as signal balancing which is not an easy task at all Specially the signal balancing becomes increasingly difficult with increase in the momber of stations. ‘Transmission media for bus LANS : ‘We can use the following transmission media for the ‘bas LANS: 1. Twisted pair 2. Baseband co-axial cable 3. Broadband co-axial cable 4 Optical fiber Advantages of bus topology : 1. The bus topology is easy to understand, install, and use for small networks. ‘The cabling cost is less asthe bus topology requires a ‘smal length of cable to connect the computers, The bus topology is easy to expand by joining two cables with a BNC barrel connector. In the expansion of a bus topology repeaters can be used to boost the sigaal and increase the distance. Disadvantages of bus topology : 1. Heavy network trafic slows down the bus speed. In bus topology only one computer can transmit and other have to wait til their tra comes and there i. 00 co-ordination between computes for reservation of transmitting time slot. ‘The BNC connectors used for expansion of the bus ‘tenuate the signal considerably. ‘A cable break or loose BNC connector will cause reflections and bring dows the whole network ee Pp 1.2.2 Ring Topology . In a ring topology, each computer is connected to the: next computer, with the last one connected to the first as shown in Fig. 1.2.3. © Rings are used in high-performance networks where lage bandwidth is necessary eg. time sensitive features such as video and audio. ‘© Bvery computer is connected to the next computer in the ring and each retransmits what it receives from the previous computer hence the ring is an active network. ae ‘cto Fig. 1.23 : Ring topology + Tae messages flow around the ring in one direction, ‘There is no termination because there i no end to the ing. * Some ring neoworks do token passing. A sbort message called a token is passed around the ring until a computer wishes 10 send information to mother computer. + ‘That computer modifies the token, adds an electronic address and data and sends it around the ring. © Each computer in sequenor reosives the token and the information and passes them tothe next computer until either the electronic address matches the address of a computer or the token retums 0 its origin. © The receiving computer retums 2 message 19 the originator indicating that the message has. been received, © The sending computer then creates another token and. places it on the network, allowing another station to capture the token and begin transmiting, ‘+ The token circulates unil a station is ready to send and capture the token. Faster networks circulate several tokens a once. + Somme ring networks have two comter-rotating rings ‘hat help them recover from network faults. Characteristics of ring LANs : + The basic ring LAN is showa in Fig. 124, which shows that along with the nodes A, B, C, D equal rpomber of repeaters are used and that the transmission is unidirectional. Networks U) ‘The data i travels in a sequential manner around the ring. Bach repeater will receive regenerate and retransmit this data bit, Problems faced in the ring topology : 1. If any link breaks or if any repeater fails thea the entire network will be disabled. “To installa new repeater for supporting a cew device, it is necessary 10 bave the identification of wo rneasby, topologically adjacent repeaters, It is necessary to take preventive measures to deat with the time jitter. Due to the closed nature of the ring topology it is necessary to remove the circulating packets, ‘These problems except for the last one can be rectified by refinements ofthe ring topology. Advantages of ring topology : 1. Every computer gets an equal acoess to the token. 2 There are no standing waves produced. 14 2 Disadvantages of ring topology : Failure of one computer on the ring can affect the ‘whole network. tis dfficul to trouble short the ring. ‘Adding oc removing the computers disnurbs the network activity. 2 3. 1.2.3. Star Topology : ‘In a star topology all the computers are connected via cables to a central location where they are all connected by 2 device called a hub as shown in Fig. 1.2.5. There is no deat ‘consections among the compoters. All the connections are snade via the central hub. + Stars are used in concentrated networks, where the endpoints are directly reachable from a central Tocation; when network expansiog is expecied and when the greater reliability of a star topology is needed. (cn Fig. 1.2.5: Star topology ‘Bach computer on a star network communicates with a central hub. The hub then resends the message 10 Layer. all the computers in a broadcast star nctwork, It will resend the message only 0 the destination computer ina switched star network. ‘The hub in a broadcast star network can be active or passive. An active hub generates the electrical signal and sends it to all the computers connected toi. ‘This type of fob is usually called a multiport repeater. Active bubs require external power supply. ‘A passive hub is a wiring panel or punch down block which acts as a connection poi. It does not amplify or regenerate te signal. Passive twbs do not require electrical power supply. Several types of cables can be used to implement a star network. A hybrid hob can use different types of cble inthe same star network, ‘A star network can be expanded by placing another star hub as shown in Fig, 1.2.6. This arrangement allows several more computers or Ihubs to be connected to that bub, This creates a hybrid star network, (6-19 Fig. 1.26 : Expansion of star topology 1.2.4 STARLANS: ‘+ In the star type LANs, the Unshielded Twisted Pair (UTP) is used asthe transmission medium, This is because the unshielded twisted pair is a Aeiephone wire which is availble in each and every office building. The other advantages of using ‘twisted wires ar as follows: 1. So no additional instalation cost is required for the installation of LAN. Since the telepbone wires cover the entire building it is possible to spread the network in every part of cach beilding, 2 Dis 1 2 ‘connected to every other device with a point to point dedicated ink as shown in Fig. 12.8. . \ Isadvantages of star topology : (6.2 Fig. 1.2.7 Single level star topology ‘The basic star topology is as shown in Fig. 1.27. This is called asa single level star topology. ‘As shown in Fig. 1.2.7, the central element ofthe star topology is an active element called hub or repeater. Each station (A. B, C, ...) is connected to the hub ‘with the help of wo links one for transmitting and the other for reception ofthe data. ‘When a single station transmits, the bub repeats the signal and sends it to each station, Typically the length of each tink is 100 m. If the twisted pair is used and the length may increase upto ‘500 mm if the optical fibre is used as transmission ‘medium. It is important to note that if two stations transmit simultaneously, then there will be a collision bbeeween their transmitted signals, If the central fub fails, the whole network fails to operate. Many star networks require @ device at the central point to rebroadcast or switch the network traffic. ‘The cabling cost is more since eables must be pulled from all computers tothe central hub. wae 1.25 Mesh Topology : In a mesh topology every device is physically ‘The term dedicated means thatthe link caries data ‘oly between two devices connected on it. A fully connected mesh network therefore has ‘(o-1)/2 physical cables to connect n devices. To Advantages : Lay accommodate that many Finks every device on the network must have o-1 inpuvoutput pots. So too many cables are required to be used for the ‘mesh topology. Node o «G-) Fig, 12.8 : Mesh topology ‘The use of dedicated links guarantees that cach connection can carry its own data reliably. ‘A mesh topology is robust because the failure of any ‘one computer does not bring down the entire network, It provides security and privacy because every message seat travels along a dedicated line. Point to point links make fault diagnose easy. Disadvantages : 12.6 Tree Topology : Since every computer must be connected to every foluer computer installation and reconfiguration is difficult, ‘Cabling cost is more. The hardware required t0 connect each link inpuvoutput and cable is expensive. A tre topology is a variation ofa star, As in asta, anodes in a tee are connected to a ceatral hub that ‘eoatrols the entire network. However, every computer is aot plugged into the ‘central hub. Most of them are connected to a secondary hub which in tum is connected to the central bub as showa in Fig. 1.29. The central hub in the tee is an active hub which contains repeater. The repeater amplify the signal ‘nd increase the distance a signal can travel. The secondary hubs may be active or passive. A passive bub provides a simple physical connection between the attached devices. 2 Disadvantages : Ifthe central hub fails the system breaks down. 1 It allows more devices to be attached to a single bub It allows the network to isolate and attach pricrtes to ‘and can therefore increase the distance of a signal ‘can travel between devices. ‘the commusications from different computers. . 2. The cabling cost is more. . 4.2.8 Hybrid Topology : Logical topology describes the manner in which the stations are logically connected to each other for the purpose of data unit exchange, Physical topology discussed earlier can be diffrent from the logical topology, of the network. . ‘As an example consider the bus topology. The bus ‘acts as 4 ceatral controller. Tt receives data and forwards itto the various nodes. “Thus the stations have a logical connection t'the bus ‘which acis as a centralized controller. ‘Therefore the logical topology of a bits is star topology, eventhough the physical topology is bus. We have discussed various basic topologies such 2s. dbus, ring, mesh, sare. Hybrid topology is the one which makes use of two ‘ot more basic topologies mentioned above, together. ‘There are different ways in which a hybrid network is created. Fig. 1.2.10 shows the hybrid topology in ‘which bus, star and ring topologies are used simultaneously. Ip Fig. 1.210, the modes 1, 2, 3, 4-md 5 ae ‘connected inthe bus topology, node 6, 7 aad 8 form star and the nodes 4,9, 10, 1, 12 are arranged in a ‘ing topology. (cay Fig. 12.10: Hybrid topology ‘The practical networks generally make use of hybrid topology. Many complex networks can be reduced to some form of hybrid topotogy. ‘The hybrid topology which is 0 be used for a particular application depends onthe requirements of that apptication. . 4.29. Hierarchical Topology : 1.2.7 Logical Topology : . In the network design for the corporate world we need to meet customer's business and ‘communication goals. For this you might need to recommend a network topology consisting of maay pieces and parts. ‘We can simplify this task by breaking things down and develop the design in pieces or layers. ‘The hierarchical network design model helps us to develop a network topology in separate layers with cach layer focusing on specific function. This hebps ‘us to choose right equipment and features for the layer. Fig. 1.2.11 shows an example of hierarchical design using routers for core, switches for distribution and ‘nubs for access. (G10 Fig. 1.2.11 : Hierarchical design using routers, switches and hubs : In Fig. 1.2.11, the high speed WAN routers camry the ‘taffic along the enterprise backbone. Then medivm speed routers convect buildings at each campus ‘whereas switcbes and hubs connect the user devices ‘and the servers within buildings. Networks growing in an unplanned way tend to develop an unstructured format. Such networks are called as fur-ball networks. ‘We can use the hierarchical topology for such fur ball networks so as io minimize the network costs ecause now we can buy the appropriate networking devices for each layer of hierarchy. 1k is also possible to accurately plan the network ‘capacity at within each layer of hierarchy. This ‘would reduce the wastage of bandwidth. Tk is possible to apply network management responsibility and network management systems to different layers to control costs. Hiecarchical topology to keep design and testing process simple, Fault isolation is improved, ‘This topology makes it possible to change the network environment, Hierarchical switch network : ‘A general hierarchical model of a network is made of three pieces. or layers as shown in Fig. 12.12, ‘The hierarchical topology is preferred because it is a sealable network. IN is easy to expand the hierarchical network. In Fig. 1.2.12, the core layer is a high speed switching and routing backbone and it should be ‘designed to pass network traffic 6s Fast as possible. ‘The distribution layer defines the network boundaries and at this layer the packet manipulation takes place. ‘The network security is also provided ot this layer. At the access layer the local end users are allowed into the network. . The layer 3 switches (coresponding to the core) are the smartest because they have to perform the routing functions calculating the best path to send traffic to destination. These switches read the network address, . ‘The layer 2 switches are less smart and slow as. compared to layer 3 switches. They can only read the data link layer (MAC) address. Layer 3 switches can ead both MAC and network addresses. (G54 Fig, 1.2.12 : Hierarchical topology using switches 1.3 Uses of Computer Networks : ‘The computer networks are playing an important role ‘in providing services to large organizations as well as to the individual coramon than. 1.3.1 Service Provided by the Network for Companies : © Many organisations have a large oumber of computers in operation. These compulers may be Within the same building, campus, city or different cites. + Benthough the computers are located in different locations, the organisations want to Keep tack of ‘inventories, monitor productivity, do the ordering and billing etc. ‘Notwories 'U) 18 © The computer networks are useful to the ‘organisations in the following ways 1. Resouree sharing : Itallows all programs, equipments and data available to anyone on the network irespectve of the physical location ofthe resource and the user. 2 High reliability due to alternative sources of data : * Tt provides high relisbiliy by having altemative sources of data, For e.g. al files could be replicated on mote than one ‘machines, so if one of them is unavailable due to hardware failore or any other reason, the ‘other copies ean be used. ‘© The aspect of high reliability is very important for miliary, banking, air waffic contol, uclear reactor safety and many other applications where continuous operations is a suust even if there are hardware or software failures. 2 Costs - © Computer networking is an important financial aspect for organisations because it saves money. © Organisations can use separate personal ‘computer one pet user insicad of using ‘mainframe computer which are expensive. © The organisations can use the ‘model {peer to poet) in which all the PCs are ‘networked together and each one cau have the access the other for communicating or sharing parpose, © The organisation, if it wants security for its ‘operation it can goin forthe domain model in which thete is a server and clients. All the clicots can communicate and acoess data through the server. 4. Communication medium : © A computer network provides @ powerful communication mediom among widely separated eenployees. * Using network it is easy for (wo ot moce employees, who 27: separated by geographical locations to work on a report, document or R and D simultaneously ie. on- lise, 1.3.2 Networks for People : ‘The computer networks offer the following secvices to an individual person: 1. Access to remote information 2. Person to person communication 3. Becommerce 4, Interactive emertainmeat, L Physical | Access to remote information : Access to remote information involves interaction between a person and a remote database. Access to remote information comes in many forms tke : © Home shopping, paying telephone, electricity bills, e-banking, on line share market etc, * Newspaper is on-line and is personalised, digial library consisting of books, magazines, scietific journals ec © World wide web shich contains information about the ets, business, cooking, government, health, history, hobbies, recreation, science, sports etc, 2, Person to person communication ‘Person to person communication includes : ‘© Blectronic-mail (e-mail) . Real time e-mail video conferencing allows remote users to communicate with no delay by seeing and hearing each other. Video-conferencing is being used for remote ‘school, getting medical opinion from distant specialists etc. . ‘Worldwide newgroups im which one person Posts a message and all other subscribers to the newsgroup can read it or give their feedbacks. 3. Interactive entertalnment : © Muliperson real-time simulation games, . ‘Video on deraand. © Patticipation in live TV programmes likes quiz, contest, discussions etc. 1.4 Social Issues : ‘New social, ethical and political problems have been faced as the computer networks have expanded in size ‘A popular feature of many networks is newsgroops ‘ot bulletin boards. Here people can exchange messages their views about atopic. But trouble comes when newsgroups discuss seasitive topics such as religion, polities or sex. ‘There are many such socal isoues associated with the computer networks, 1.5 Network Hardware ; ‘Now let us discuss the techaical isues involved in the network design. ‘Two important dimensions of a computer network ae: 1. "Transmission technology and 2, Seale. i I Networks ( 4.5.1. Types of Transmission Technology : The transmission technology can be categorised broadly into two types : 1. Broadcast aeiworks and 2. Poimt-to-point networks, 1. Broadeast networks: Ina broadcast networks all the machines on the network use or share communication channel that is shared or used by all the machines on the network. Short messages called packets sent by ‘any machine are received by al the others. ‘+ Broadcast systems generally use a special code in the address field for addressing a packet to all the concemed computers. This mode of opetation is called broadcasting. + Some broadcast systems also. support transmission to only a group of few machines ‘known as multicasting, ‘+ When a packet is received, a machine checks the address field. If the packet is addressed to it then the [Link] processed, otherwise the packet is ignored. 2. Point-to-point networks : In point to point networks there exist of many connections betwee individual pairs of machines. "To go from the source to the destination a packet on this types of network ‘may have to go through ialermediate computers before they reach the desired computer. ‘© The packets cmenging from the same source have to follow multiple routes, of diferent lengths. © Hence properly designed routing algorithms are . very important in te point-o-point networks. * An ignportant general rue is as follows : PU) 1.6 Network Scale ‘* This is an alternative critetion for classification of networks. ical + Fig. 1.6.1 gives the network classification based on their physical size, All these systems are multiprocessor ; ee Eiample of +, Retwrark 01m Data flow board machine Tm ‘Same system | Muti 10m Same room LAN 100m ‘Same LAN building Ten ‘Same TAN ‘campus 1k ‘Same ci MAN 100 kn ‘Same state ‘WAN. 1,000 ken ‘Same WAN continent 10, 000 km ‘Same plaset_| Internat Fig. 1.6.1 : Network classification according to scale © Beyood the multicoraputers are the true networks, in which the computers communicate by exchanging messages over long cables. © Such . networks are divided into following categories : 1. Local area networks 2. Metropolitan networks and 3. Wide area networks. Internetwork : '* The connection of two or more networks is called as ‘an internetwork. © The best example of intemetwork is the Internet. 1.7 Network Classification by their Geography : "| © Computer network can be classified based on the ‘geographical area they cover, ie. the area over which the neework is spread. © Such a classification is shown in Fig 1.7.1. . In this section, we will discuss the following catogories of networks: PAN uN MAN WAN CAN (Personal area ——_(Looal area (Metropottan (Wide area (Campus area network) ‘notwork) area network) ‘notwork) ‘network) (G-1400Fig, 1.7.1 : Network categories Networks 410 4.7.1 Local Area Networks (LAN) : ‘© The Local Area Network (LAN) is a network which is designed (o operate over a small physical area such as an office, factory or a group of buildings. LANs are very widely used in a variety of applications. © LANs are easy to. design and troubleshoot, The personal computers and workstations in the offices: are interconnected via LAN. ©The exchange of information and sharing of resources becomies easy because of LAN. © In LANall the machines are connected to a single cable. Diffeeat types of topologies such as Bus, Ring, Star, Tree ete. are used for LANs. © LAN uses a layered architectore and they are capable of operating at hundreds of Mbits/sec. © A Local Area Network (LAN) is usually a privately ‘ovmed and Tinks the devices in a single office, ‘building or campus of upto a few kilometres in size as shown in Fig. 1.7.1. Depending on the needs of an organisation and the type of technology used, a LAN can be as simple as a few computers and a printer at home. or it can contaia ‘many computers in a company and include voice, ‘sound and video peripherals. © LANs are widely used to allow resources 10 be shared between personal computers or workstations. ‘The resources 40 be shared can be hardware like a ‘printer or softwares or data, © Ina LAN one of the [Link] become a server serving all the remaining computers called clients. Software can be sited on the server and it can be ‘used by the remaining clients, . LAN’s are also distinguished from MAN's and WAN’s based on the transmission media they use and topotogy. In general a given LAN will use only ‘one type of transmission medium. The most common networking topologies used are bus, sing and star. . ‘The data rates for LAN can now range from 10 Mbps 1 16 Gbps, Tnnportant characieristics of LAN : 1. Very high degree of interconnection between the computers. High rte of data wansmission. Physical connection of computers in a LAN is easy. very computer in the LAN can communicate with Aer every other computer. 5. The medium wed for data transmission is ‘ncxpeasive, LAN topologies : ‘Various topologies are possible for the broadcast LANs such as bus topology or ring topology as shown in Fig. 17.2. (b) Ring topology (ean Fig. 172: LAN topologies Static and Dynamic broadeast networks : + The broadcast networks are further classified into ‘wo types aamely : 1. Static networks and 2. Dynamic networks. © This classification is based on how the common ‘chanel is allocated, In static allocation, each machine is allowed to broadcast ooly in its alloted time slot. © But static allocation wastes the channe) capacity when & mavhive does not want to transmit in ite allotted time slot, ‘+ Hence most of the systems ty to allocate the channel dynamically ic. on demand. LAN coaponents : Some of the important LAN components are as follows: 1. Workstations, 2, File servers. 3. Gateway. 4, Network interfacing uait, 5. Active and passive hubs. 6 LAN cables or commonication channels, Workstation : ‘Workstation refers to the individual, single computer. ‘A communication capability is added to enable it for vetworking. File server : File server is a computer that allows the sbaring of data, software and hardware resources by runing special softwares. ‘Computer Networks Gateway: It assists the transfer of data from one LAN to the other LAN, =~ [Network Interfacing Unit (NIU): 1k is a unit which consists of hardware as well as software. I uses microprocessor to control the access and communication ina network, LAN cables or communication channel : A cable is used for comecting the computers in a LAN. The communication from one computer to others takes place over the cables. So cables are called communication channels. The twisted pat, coaxial cables or optical fiber cables are used in LANs, Advantages of LAN : 1. High reliability. Faitue of individual computers does ‘ot affect the entire LAN, 2. Iki possible to add a new computer easily, ‘The transmission of data at a very high rate. Sharing of petiphetal devices such as printer is possible. Applications of LAN : File transfer and file acess. Personal computing, Office automation. Distributed computing. Word and text processing. Document distribution, Remote access to databese. 3. _Hlectronic message handling, 1.7.2. Ethernet: © — Both Internet and ATM (Asynchronous Transfer ‘Mode) were designed for wide area networking. But ay Nae eee ‘in magy applications, a large number of computers are 10 be connected to each other. + For this the Local Area Network (LAN) was inwoduced. The most popular LAN is called Ethernet. ‘+ The IEEE 8023 standard is popularly called as Ethemet. It is a bus based broadcast network with decentralized control. + Itcan operate at 10 Mbps or 100 Mbps or even above 1 Gbps. + Computers on an Ethernet can transmit whenever they want to do so. If two or more machines transmit © Then the transmitting computers just wait for an arbitrary time and retransmit their signal. ‘+ There are various technologies available in the LAN market but the most popular one of them is Ethernet. © Im this section we ate going to discuss three ‘generations of Ethernet: 1. Traditional Ethesnet (10 Mbps) 2, Fast Rthernet (100 Mbps) 3. Gigabit Bthernet (1000 Mbps) © Traditional Ethemet was created in 1976 and has a data rate of 10 Mbps. © The fast Bthernet is its mext version and has a data rte of 100 Mbps. * The Gigabit Ethemet ‘operates at the data rate of 1000 Mbps or 1 Gbps. Why is it called Ethernet ? This system is called as Bthemet after the Iuminiferous ether through which the electromagnetic radiation was once thought to propagate. (cassFig. 1.7.2): Architecture of original Ethernet g Senet hemor Soap srr) Physical Layer ‘Transmission medium ‘The transmission medium is thick co-axial cable (Called ether) upto 2.5 km fong. Repeaters are placed fer every 500 meters. Upto 256 machines can be attached to the maultidrop cable. ‘The architecture of the original Btheret is shown in Fig. 17.20. The original Bthemet was standantized as IEEE 802.3 standard. The committee also standardized token bus (802.4) and token ring (802.5) standards ‘which were aot as popular as Ethernet. 1.7.3 Metropolitan Area Network (MAN) : A MAN js basically a bigger version of a LAN and ‘sormally uses similar technology. It is designed to extend over a larger area such as an entire city. The MAN can be in the form of a single network such as a cable network or it can be a combination of ‘muiple LANs as shown ia Fig. 1.733. ‘A MAN may be wholly owned and operated by a Private company or it may be a service provided by a public company, such a$ a local telepbone company (telco). ‘A MAN is distinguished by the TEBE 802.6 standard oF it is also known as Distributed Queue Dual Bus (0QDB). The DOQDB consists of two unidirectional cabtes (buses) to which all the computers are connected as shown in Fig. 1.74. Each bus bas a device which initiates the ransmission activity called as the head-end. ‘Traffic that is destined fo 2 computer to the right of the sender uses the upper bus aad to the left uses the lower bus as showa in Fig. 1.74. «G-0Fig. 1.7.3 : Metropolitan area network Dion o ta fow on Dus A—+ (cs0Fig, 1.74 : Distributed queue dual bus architecture(DQDB) g Computer Networks (Comp-SPPU) Physica Layer 1.7.4 Wide Area Network (WAN) : ‘When a network spans a large distance or when the ‘computers t0 be connected to each other are at Widely separated locations a local area network ‘cannot be used, For such situations a Wide Area Network (WAN) most be installed. The communication between different users of “WAN” is established using leased {elephone lines or satelite Tinks and similar cheinels, This cheaper and more efficient to use the phone network for the links. ‘Most wide area networks are used for transfering large blocks of data between its users. As tbe data is from existing records or files, the exact time taken for this data tansfer isnot a citcal parameter. ‘An example of WAN is aniline reservation system. Terminals are located all over the country through which the reservations can be made. ‘tis important to note here that al the terminals use the same centralized common data provided by the ‘central reservation computer. Because of the large distances involved ia the’ wide ‘area networks, the propagation delays and variable signal travel times are major problems. ‘Therefore most wide area nétworks sre not used for time critical applications. As explained earlier they are more suitable for transfer of data from one user to the other which is aot a time critical application, ‘Wide area networks are basically packet switching nesworks. ‘A WAN provides long distance transmission of data, voice image and video information over large ‘geographical areas that may comprise a county, a ‘continent or even the whole world as shown in Fig. 1-75. WAN contains a collection of machines used for running user (te. application) programs. All the machines called hosts are commected by a communication subnet as shown in Fig. 1.7.6. ‘The function ofthe subnet is to camry messages from hhost to host. The subnet conssis of two important components; transmission lines and switching elements. Transmission lines move bits from one machine to another. The switching elements are specialised computers used to connect two oF more transmission ines. When data arrive on an incoming line, the switching element bas to choose an outgoing line on which itis to be forwarded, ‘©The switching clements are cither called as packet ‘switching nodes, infermediate systems, data switching exchanges or routers. . ‘When a packet is sent from one router to another via roviors, the packet is reosived at intermediate router. It is stored in the ‘outers until tbe required output line i fee and then forwarded. A subnet using this principle is called & point to point, or packet switched subnet. cal area. A WAN that is wholly owned and ved by a siagle company is offen called 2s an enterprise network, © In most WANS the network contains a large number ‘of cables or telephone lines each one coonecting @ pair of routers. '* fewo routers which are not connected to each other via a cable want to communicate, then they have to do it indirectly via oer rouers. Router interconnection topologies = + Fig. 17.7 shows some of the possible router intzrconsection topclogies ina point to point subuet. *O {a)Stor sn () Ring (Tree «osn Fig, L7.7 (Contd..) Coneunicaion surat (¢-s6Fig. 1.7.6 : Communication subnet and hosts soon Fig. ‘The LANs have a symametic topology witle WANS have irregular ‘The WANs can also be formed using satellite or round radio system, Satellite networks are inherently broadest type 60 they are useful when the ‘broadcast property is important. ‘Characteristics of WAN: Following are some of the important characteristics of WAN: 1. Remote data entry and access is possible, 2, Communication facility is provided. 3. Centralized information is created and used. 1.7.5 PAN (Personal Area Networi) : 'A Personal Area Network (PAN) is a computer network designed for and organized around an individual person, ‘A PAN generally consists of a mobile computer such as a laptop, a cell phone and for a Persoaal Digital ‘Assistant (PDA), PAN will allow the communication to take place among these devices. PAN can also be used for communication among personal devices themselves (intrapersonal communication) or for connecting to a higher level ‘network and internet (This is called a an uplink), Networks ( 1 ‘The PANs can be constructed using cables or it can bbe wireless. The wireless PANs typically use Bluetooth or sometimes use the infrared connections, ‘The PANs generally cover a range upto 10 meters. PAN can be considered as a special type of Local ‘Area Network (LAN), which is designed for one person instead of a group. 1.7.6 CAN (Campus Area Network) : ‘The Campus Area Network (CAN) is made up of an interconnection of LAN within a litited geographical area. ‘The network equipments such as switches, routers and the transmission media i.e. optical fiber ete. are almost entirely owned by the campus owner (ie. a ‘company, university, government etc.) For example, a university CAN would connect different buildings in its campus, such as vations departments, library, student hall to cach other. CAN could also be thought of as a special case of WAN. 1.7.7 Wireless Networks : ‘The fastest growing segment of the computer industry is the mobile computers such as aotebook computers and Personal Digital Assistant (PDAs). ‘The wireless networks are becoming. increasingly important because the wired connection is not possible in cars or aeroplanes. Wireless networks can have many applications. A. ‘very common example is the portable office, People travelling on road often want to make use of their portable electronic equipment foc telephone calls, e-mails, faxes, read remote files ete. Physical Layor Wireless networks can exist om trucks, buses, taxies, ‘aeroplanes etc. They are used where the telephone systems are destroyed in the event of disasters such as fires, floods and earthquakes ete, ‘The wireless networks are important for military. Wireless networks and mobile computing are related bur they are not idcatical because portable computers are sometimes wired and some wireless computers ate not portable But some applications are truly mobile wireless applications such as a portable office, inventories being bandled by PDAs, etc. Wireless LAN is another example of wireless network, Direct digital cellular service CDPD (Cellular Digital Packet Data) is now becoming available. It is possible. to have combinations of wired and wireless networking 1.7.8 Internetworks : ‘When two or more networks are connected together they are called as intemetwork or intemet as shown, in Fig. 178. Individual networks are joined into imernetworks by the use of intemetworking devices like bridges, rovters and gateways. Fig. 17.8 shows a general form of intomet. It is the collection of number of LANs which are ‘imverconnected via a WAN. ‘What is the difference between a subnet and WAN? JF the system within a closed periphery contains only routers then it is called as a subnet, But if it contains routers a well as hosts then itis a WAN. 1.7.9 _ Com; Networks (Comp-SPPU) ison of LAI WAN and MAN 1. (@) Ad-hoc mode (Gasen Fig. 18.1 : Types of 892.11 networks ‘The otber mode of operation is ad-hoe networks as shown in Fig. 18.1(0). In this mode a group of computers ean communicate to each other directly ‘without any Access Point (AP) in between, Ownership of network | Private public ‘Geographical Area | Smalj ‘Very large (states or covered comntries) Design and | Easy Noteasy Not easy maintenance ‘Communication ‘Coaxial cable PSTN or satellite | Coaxial cables, PSTN, optical sedi Tinks fitecables virekss "| { Data rates (speed) High low Moderate ‘Mode of Each station can transmit | Bach station cannot | Each station can transmit or. communication and receive transtnit receive. Principle Operates on the principle of | Switching Both broadcasting Long Moderate High Moderate . The infrastructure mode is as shown in Fig. 1.8.1(@). The client such as laptop ot a smart phone is connected to another network such as company Intemet. . In this mode the client is associated with an Access Point (AP) which is intum connected to the other network as shown. + The client sends and receives its packets via AP. ‘Many such APs are connected together to form an, extended 802.11 network. 9 Network Classification by their Component Role — nee ‘The focal area networks are classified into two types: 1. Peer co peer networks. 2. Client server networks, ‘The relationship between each PC or device on the network with the others in terms of control will be ‘dependent on the choice of network type. For these two types, the special software is required for controlling the flow of information between the sets. ‘The Network Operating System (NOS) is installed ‘on each PC depending ott the type of network. NOS monitors the data exchange, How of files, and other ponitors ‘The network operating systems are different for the eer to peer and client server networks. A peer-to-peer network is analogous to a company that uses decentralized management, where decision are made locally. r Networks + A clicat-server network is similar to company that ‘works on the principle of centralized management, ‘where decisions are made in a central location. 1.10 Peer-to-Peer Networks : © Fig. 1.10.1 shows the stucture of the peer-to-peer network. In this type of network, each computer is responsible for making its own resources availabe to ‘other conaputers on the network. © Esch computer is responsible for setting up and _maintaining its owa security for its resources. + Also each computer is responsible for accessing the Tequired network resources from peer-to-peer relationships, * Peer to peer network is useful for a small network containing less than 10 computers on a single LAN, Each computet maintains its own accounts and their security settings. © Ta peetto-peet network, every computer can function as both a clieut and server. Windows 2000 ‘comes in both server and professional versions, but it’s sill a peer-to-peer operating system. * Peer to peer networks do not have a central contol system. There are no servers in peer networks, + In this type of network users simply share disk space ‘and resources, such as printers and faxes. * Peer networks are organised into workgroups. \ Workgroups have very litle security. There is no ‘central login process. ‘+ If the user has logged into one peer on the network ‘he can use any resources on the network that are not ‘controlled by a specific password. Poor Peer a —s en Peer Poor (6-oFig, 1.10.1 : Peer-to-peer network relationship © Access to individual resources can be controlled if the user who shared the resources installs a password to access it. + Since there is tid central security, the user will have ‘to know iidividual password foreach secured shared resource which hé wants to access, * Peer to peer nétirorks are relatively simple. Each ‘computer inthe Htwork can act as client as well as server ad per requifement. ‘This elithinates i aced of expensive server. No additional software is necessary in order to set up the peett8 peer network. 1.10.1. When t6 Ue Peer to Peer Networks ? ‘The peer i8 peer networks are suitable for the following working conditions: ‘* _fnetwork security is not an important issue, * If the nifiber of users is less than #0. (small setwork). ‘+ fall be users are situate in the same area, * Ifthe possibilty of future expansion is less, 1.10.2 Advantages of Peer to Peer Networks : Peer networks have many advantages, especially for ‘stall business houses that cannot afford to buy expensive server hardware and software, 1, Nocera investment in server hardware or software is required. 2 Useless expéisive computer hardware : In peerto- Peet network, the resources are distributed over ‘many computers, so there is no need for higher-end- server computer. 3. Easy to administer : In peer-to-peer network each ‘machine performs its own administration, 4 No NOS required : Peer-to-peer network does not require a Necwork Operating System (NOS). 5S. More builtmredundancy : If you have a small network, with 10-20 workstations and each one with ‘some important data on it, and onc fails you still hhave most of your shared resources availabe. Peerto-peet network/achieve more redundancy because of smaller possibility of single point of failore. Easy setup and lower cost for small networks, ‘Users can control resource sharing. ‘A oser is not dependent on other computers for its ‘operation. 1.10.3 Disadvantages of Peer to Peer Networks : There are several disadvantages of peer-to-peer network, particularly for larger networks as follows : ee g@ ‘Computer Networks (Comp-SPPU). Phy 1. Individual performance is affected : If some workstations have fequendly astd resources. on them, then the use of CSE resources by other ‘computer might adversely A@fects tHe person using (his particular workstation. 2. Less security : A peer-to-peer nctwork operates on the most common desktoy’ opetatnig. systems like windows which are not #éfy secure operating systems, 3. Backup is dficule : In peer-to-pedt network there is ‘Ro centralized server, Hence data is scattered over sany workstations. So it is diffiealt to backup all ‘data in an organized manner. 4. Hard to maintain version control ; [a peer-to-peer network, files are stored on number of different workstations. So it is difficult to manage different document versions or fies. As there is no ceatralived management it makes large oer networks hard to manage and final data easily. Users are supposed to manage their own computers. It isnot possible to save important date in a centralized manner. ‘Additional load on computer Vetause of resource sharing and absence of server. 1.11 Client / Server Network (Server Based Network): In client-server network relationships, certain computers act as server and other act as clients. A server is simply a computer, that makes the network resources available and provides service to otter ‘computers when they request i A cliemt is the computer running a program that requests services froma server. * Local Arex Networking (LAN) is based on the clent- server network relationship. You can construct a Client server network by using one or more powerful networked computers as a servers and the rest of 35 liens, Client-server network typically uses a directory service to store information about the network and its users. ; Acdlient-server nctwork is one in which all available network resources such as files, directories, ‘applications and shared devices, are centrally managed, stored and then are accessed by client. Fig. 1.1111 shows client-server network relationship. In the client server networks the servers provide security and administration ofthe entre network, © In client-server networks the processing tasks are vided between clients and servers. Clients request services such as file storage and printing and servers deliver them. Client ‘The individual workstations in the network are called asthe clieas. Server: © The ceatral computer which is more poweeful than the clients and which allows the clients to access its softwares and database is called as the server. + Sexier computers typically are rnore. powerful thas client computers or are optimised to function as servers, No user can access the resources of the servers until fhe has been authenticated (permitted) by the server fo dos0. USS Gront ‘Ghent ‘Giont| ‘Ghent ‘o-anFig, LIL : Client server network relationship ¢ Networks (Comp-SPI 1.11.1 Communication in Client-Server Configuration : . . 1.11.2 explains the principle of communication ins cea secon, © The client places a request on the server machine ‘when be wants an accesso the ceatralised resources. (GaniFlg. 1.11.2 : Clentlserver communication ‘+ The server responds to this request and sends the signal accordingly to the client as shown in Fig. 111.2, +The software run at the client computer is called as clieat program. This software configures. that * Similarly the software run on the server computer is called as server program. It configures that particular ‘COMPULEE 10 act as a server. 1.11.2 Advantages of Client-server Network : The advantages of client-server network are as follows : 1. The network is secure : In client-server network's high security is because of |F several things : (@) Shared resoues are located in a centraized ‘area and they are sdministered ceatrally. (©) The servers are physically placed in secure location such as lockable separate server room, (©) The operating system runs on client-server are designed wo provide beter security to network. @ Better security to network due to good administration, 2 Better performance : ‘The dedicated server computers sre more expensive than standard computer workstations, but they also offer considerably better performance. 3 Centralized backup : ‘Backing up company's important data is much easier when it is located on a ceawalized server. Centralized backup is much faster too. 4. Higher reliability : In clicat server network centralized dedicated server provide more reliability. i bas buil-in redundancy, 5. Central file storage, which allows all users to work from the same of dita, 6. Reduces cost because of sharing of hardware and software, Increased speed dive vo dedicated server for sharing resources. 8. Single password allows aocess to all shared resources, 9. Central organisation which keeps data from getting lost among computers and easy manageability of large number of users. 10, The individual users don’t have 10 manage or share resources, 1.11.3 Disadvantages of Client-server Networks : 1. Professional administeation is required : Client- server networks usually need professional administration. You can hse a network administrator ‘or you ean use a company which provides professional network administration services, 2. We have to use a high speed server computer with dots of memory and disk space. 3. lesequires a special network operating system and 2 taurber of cient licenses, 4. Expensive dedicated hardware needs t be used. 1.11.4 Applications of Client-server ‘Configuration : ‘Some of the important applications are 2s follows : © Bemail clients, © Web browsers, * FIP (ite transfer) clients. 11S between Peer-to-Peer Network and Client-Server Network : Hi a Tis much like company { It is much like company | decentralized | using centralized ‘Not built-in redundancy, its sabe Re ae] network. Better performance, Networks 1-20 1.11.6 Distributed Networking : Distributed networking is a distributed computing network system. Its sid to be distributed when the competer programming and the data to be worked on ‘are spread out across mote than one computer. Usually the distributed networking is implemented over acompuer network, Prior to the emergence of low cost desk tops, the computing was generally centralized to one ‘compoter. Although such centers still exist, the distributed networking applications operate more efficiently. ‘A popular example of disttibuted networking isthe cient server network. However very large enterprises find it extremely difficult to manage their distributed netviork using, he traditional client server approach. So the recent evelopment in the field of clood computing has ‘opened up new possibilities. We may define the distributed network as a type of computer network that is spread over different networks. This provides a single data coranumicuion network, which can be managed jointly or separatcly by each network, Besides shared communication within the network a Aistributed network flew uses the distibuted processing. Distributed networks are part of distribaled ‘computing architecture, in which enterprise IT infrastructure resources are divided over a number of networks, processes and intermediary devices. A distributed network is powered by a network management software, which manages and monitors date routing combining and allocating network bandwidth, access control and other important networking procestes. Fig. 1.11.3 shows the simplified Distributed Network ‘Architecture (DNA). 1.11.7 SDN (Software Defined Network) : In the recent years, there has been a wemendous increase in casrier network traffic. This is basically due to the explosive growth in the use of online applications and inthe mobile connected devices, Now a days the network operators need to cope up ‘with a vast variety of data formals, service types and online deviees. ‘They need 0 do so while ensuring security, quality and availability without increasing the costs. However the existing network architectures and theit ‘management tools are not designed to cope up with such highly elastic demands, The Software Defined Network (SDN) is the industry's response to meet all the challenges nieationed above. SDN allows the networks to react dynamically tothe ‘changes such as usage pattems and availability of network resources, At is possible to instantly adjust the network atchitectue in response to user or application request sv that sotvices can be introduced very easily, dguickly and at lower cost, In SDN a separation is provided between the controt plane (controller) and data plane (switch) functions ‘of networks with the belp of a protocol that modifies the forwarding tables in nerwork switches, ‘Doe to this the networks can be easily optimized so that they can respond quickly to the changes in the network usage, without manvally reconfiguring the existing infrastructure or hardware. SDN also provides an entity ie. the controler with which the switches and applications can ‘communicate in realtime, Due to the controler itis possible forthe networks to interact with applications. This allows’ the applications to implement multiple fogical necwork topologies on a single network fabric. : (cre Fig, 1.11.3 : A distributed video surveillance network ‘Computer Networks Open flow : + It is a multivendor standard defined by the Open Networking Foundation (ONF) for implementing SDN in networking equipment. The open flow protocol defines the interface between ‘an opea flow controller and an opea flow switch. . ‘The open flow protocol allows the controller to instrict the switch about handling of incoming data packets. How does SON work ? © SDN provides a wide range of competing architectures but a its most simple, the SDN method centralizes control of network by separating the ‘contro! logic to off device computer resources. + All SDN’ models tave some version of an SDN controller as well as southbound APIs and north bound APIs. 1. Controliers : Controtter isthe brain of SDN networks. It offers a ‘centralized view of the overall network and enables the network administrator to dictate the underlying systems (such as switches or routers) about the traffic handling. 2.“ Southbound APIs: . SDN uses southbound APIs to relay information to the switches and routers “below”, Southbound APLis ‘one of the most common protocols. “Open Flow” considered as the first standard in SDN was the first. southbound APT. © Some people consider open flow and SDN as one and the same. But acully open flow is one piece of SDN. 3. Northbound APIs : . SDN uses northbound APIs to communicate with the applications and business logic “above. . ‘This helps the network administrators to shape traffic and deploy services. The SON frame work: Fig. 1.124 shows the typical SDN framework. The ‘SDN architecture bas the following important features. ‘Features: e Directly programmable Agile Centrally managed Programmatically configured ‘Open standard based Vendor aevtral aaa «G19 Fig. LILA: SDN framework: 1. Directly programmable : In SDN the network contre is directly programmable because itis decoupled from the forwarding functions. 2 Agile: As the control is decoupled from forwarding the administrators dynamically adjust network-wide traffic flow to meet the changing needs. 3. Centrally managed : In SDN controllers the network intelligence is centralized which maintains a global view ofthe network. It appears to the applicaions and policy engineers as 2 single Togical switch. 4, Programmatically'configured ; SDN allows the network managers to configute, manage, secure and optimize the network resources very quickly with the help of dynamic, automated SDN programs, 5, _ Open-standard based and vendor neutral : ‘As SDN is implemented through open standards, the ‘network desiga and operation get simplified. SON is not same as open flow : + Many a times people consider open flow same as SDN. But it is not wue. As shown in Fig. 1.11.4, open flow is just one element of the SDN architecture, + Actually open flow is an open standard for communications. projocols that enable the contol plane to interact with the forwarding plane. . ‘Note that open flow és not the only protocol available for SDN. f Networks (Comp SI 122 Benefits of SDN: ‘Nis direely programmable, thas a centralized management. Delivers agility and flexibility. Enables innovation, Itis dynamic and manageable. Itis cost effective. Is ideal for high bandwidth, dynamic applications ‘SDN applications : |. Software Defined Mobile Networking (SDMN). 2 SD-WAN : It is a WAN managed using the principle of SDN. SD-LAN : Tt is a LAN that works on the principle of ‘SDN. 12 Layered Tasks ‘The concept of layers is used in our daly life. Take an example of two fiends with one friend wants to send a gift 10 the other via courier service Fig. 1.12.1 shows the steps involved inthis process. In Fig, 1.12.1, we have three important persons involved namely the sender, the receiver and the cartier who caries the gift box, from one city to the other. Hierarchy of tasks : ‘The point to be noled is that in order to complete a task in day today life small actions are being done in a hierarchical way or layered manner. See ee 3 1. Atthe sender: ‘The tasks of higher layers : 1. Get the gift item 2. Pack itin abox 3. Write the destination address on the box. ‘Middle Inger : Carry the adressed box tothe office of a courier company. Lower layer + Give the box 1o a person who will take it to the destination city. At the receiver: ‘Tasks of lower layers : The box is delivered to the ‘courier company office in the destination city. ‘Middle layers : The bor is carried by another person to the destination address and the box is delivered, ‘Upper layers = 1, Receive the box 2 Unpack it 3. See the gift Hierarchy and layered tasks : ‘This discussion demonstrates thatthe important tasks ae carried out by the higher layers whereas the simpler tasks are carried out by the middle and lower layers. Tn the network protocols as well the layered architecture is used, Box carted em cours Todestnatr (G- Fig. 112.1: Layered tasks 1.13_ Network Software as the hardware. The network software is highly structured now a days, 1.13.1 Protocol Hierarchies (Layered ‘The software sed in networks is equally important Architecture) : Most networks ate organized in the form of a series of layers or levels as shown in Fig. 1.13.1. This reduces the design complexity, ‘The number of layers, the mame of each layer, the Contents of each layer and the function of each layer

You might also like