 Application Layer: Overview

 Network Application Examples

 Creating Network app

 Network Application Architectures

 Client-Server Architecture / Peer-to-Peer Architecture

 Process Communicating
 Client and Server Processes
 The Interface Between the Process and the Computer Network – Socket
 Addressing Processes

 Transport Services Available to Applications:

 Reliable Data Transfer, Throughput, Timing, Security

 Transport Services Provided by the Internet

 TCP Services / UDP Services

 Application-Layer Protocols
Chapter 2: Application Layer
Book: Computer Networking: A Top Down Approach.
Author: Jim Kurose, Keith Ross Addison-Wesley
Our goals:  learn about protocols by
 conceptual and examining popular
implementation aspects application-layer
of application-layer protocols
protocols • HTTP
• transport-layer service • SMTP, IMAP
models • DNS
• client-server paradigm  programming network
• peer-to-peer paradigm applications
• socket API
  Since 2000
• voice and video applications
• voice over IP (e.g., Skype)
 Internet applications • video conferencing over IP such
• classic text-based applications as Skype
• became popular in the 1970s • user-generated video
and 1980s distribution
• text email, remote access to • streaming stored video
computers, file transfers, and (YouTube)
newsgroups • movies on demand such as
• text messaging Netflix

 Include the World Wide Web  multi-user network games

• Encompassing surfing, Search,  real-time video conferencing
Electronic Commerce
 social networking
 include (Facebook, Twitter, WhatsApp)
• instant messaging and
 …
• P2P file sharing
 application

At the core of network

transport
mobile network
network
data link

application development is physical

national or global ISP

• write programs that:

• run on (different) end systems
• communicate with each other over
the network
• E.g. Web application local or
regional
• two distinct programs that communicate ISP
with each other
home network content
• browser program running in the application
provider
transport
• user’s host (desktop, laptop, tablet, network network
application
datacenter
network
smartphone….) data link
physical
transport
network
• Web server program running in data link
physical
• Web server host
• e.g., web server software enterprise
network
communicates with browser sofware
 application

 when developing your new

transport
mobile network
network
data link

application physical
national or global ISP

• need to write software that will

run on multiple end systems
• for example, in C, Java, Python
 no need to write software local or
regional
for network-core devices ISP

• network-core devices do not home network content

application
provider
run user applications transport
network network datacenter
application
network
• applications on end systems allows data link
physical
transport
network
for rapid app development, data link

propagation
physical

enterprise
network
 application’s architecture is distinctly different from the network
architecture (e.g., the five-layer Internet architecture
 From the application developer’s perspective,
 the network architecture is fixed and provides a specific set of services to
applications.
 The application architecture, on the other hand,
 is designed by the application developer and
 dictates how the application is structured over the various end systems.

 Two application architecture

 Client-Server architecture
 peer-to-peer (P2P) architecture
 two factors are involved :
 Servers
• A server is the one who provides requested services.
• the server hosts, delivers, and manages most of the
resources and services requested by the client.
• Types:
• Mail servers:
• File servers
• Web servers:
 Clients It is also known as the networking computing
• Clients are the ones who request services. model or client server network as all
requests and services are delivered over a
network.
 For example, in hospital data processing,
• a client computer can be busy running an application
program for entering patient information, meanwhile
• the server computer can be running another program to
fetch and manage the database in which the information is
permanently stored.
server:
 always-on host
 permanent IP address
• Because the server has a fixed,
• well-known address, and
• the server is always on
 often in data centers, for scaling
 services requests from many other hosts
• clients:

• clients do not directly communicate with each other

• Web application,
• two browsers do not directly communicate
• contact, communicate with server
• may have dynamic IP addresses
• examples: Web, FTP, Telnet, email
 two or more systems are connected together
and they share resources without the need of a
central server.
 (P2P) architecture consists of a decentralized
network of peers –
• nodes that are both clients and servers
 Some uses of P2P architecture:
• File sharing
• Instant messaging
• Voice Communication
• Collaboration
• High Performance Computing
 Some examples of P2P architecture:
• BitTorrent
• Skype
• Bitcoin
 no always-on server
 application exploits direct communication between pairs
connected hosts, called peers
 peers are not owned by the service provider mobile network

 arbitrary end systems directly communicate national or global ISP

 peers communicate without passing through a dedicated

server,
• the architecture is called peer-to-peer.
 peers request service from other peers,
provide service in return to other peers local or
regional
• self scalability – new peers bring new service ISP
capacity, as well as new service demands home network content
provider
 cost effective, network datacenter
network
• don’t require significant server infrastructure and server
bandwidth
 peers are intermittently connected and change
IP addresses enterprise
network
• complex management
 example: P2P file sharing
process: program running within a clients, servers
host
client process: process that
 within same host, two processes initiates communication
communicate using inter-process server process: process that
communication (using rules
waits to be contacted
defined by OS)
 processes in different hosts
communicate by exchanging
messages  note: applications
with P2P
 With the Web, a browser is a
client process and a Web server is
architectures have
a server process. client processes &
server processes
 process sends/receives messages to/from its socket
 A process is analogous to a house and
 socket analogous to door
• sending process shoves message out door
• sending process relies on transport infrastructure on other side of door
to deliver message to socket at receiving process
• two sockets involved: one on each side
 to receive messages,  identifier includes both IP
process must have address and port numbers
identifier associated with process on
 host device has unique 32- host.
bit IP address  example port numbers:
 Q: does IP address of host • HTTP server: 80
on which process runs • mail server: 25
suffice for identifying the  to send HTTP message to
process? gaia.cs.umass.edu web server:
 A: no, many • IP address: 128.119.245.12
processes can be • port number: 80
running on same
host
• Process-to-Process Communication using socket addresses

24.15
  protocols are usually designed
 types of messages by
exchanged, • companies/organizations for
serving a particular purpose
• e.g., request, response
open protocols:
 message syntax:
 defined in RFCs, everyone has
 what fields in messages & access to protocol definition
how fields are defined  allows for interoperability
 message semantics  e.g., HTTP, SMTP
• meaning of information in proprietary protocols:
fields • usually designed by manufacturers
for usage with their own products
 rules for when and how only
processes send & respond to • e.g., Skype

messages
 Many networks, including the Internet,
 provide more than one transport-layer protocol.

 When you develop an application,

 you must choose one of the available transport-layer protocols.

 How do you make this choice?

 Most likely, you would study the services provided by the available transport-layer
protocols, and then pick the protocol with the services that best match your application’s
needs

We can broadly classify the possible services along four dimensions:

• reliable data transfer,
• throughput,
• timing, and
• security.
 packets lost within a computer network
• overflow a buffer in a router
• discarded by a host or
• router after having some of its bits corrupted
 guarantee
• data sent by one end of the application is delivered correctly and completely to the
other end of the application
• transport-layer protocol can potentially provide
• to an application is process-to-process reliable data transfer

Reliable data Transfer

 some apps (e.g., file transfer, web transactions) require 100% reliable data
transfer
 other apps (e.g., audio) can tolerate some loss
 transport-layer protocol can potentially provide to an application is process-to-
process reliable data transfer
timing guarantees:
many shapes and forms
every bit
• that the sender pumps into the socket arrives at the receiver’s socket
• no more than 100 msec later.
• interactive real-time applications
 some apps (e.g., Internet telephony, interactive games) require
low delay to be “effective”
throughput
 guaranteed available throughput at some specified rate
 some apps (e.g., multimedia) require minimum amount of throughput to
be “effective”
 bandwidth-sensitive applications
 bandwidth-sensitive applications have specific throughput requirements, .
 other apps (“elastic apps”) make use of whatever throughput they get
 Electronic mail,
 file transfer, and
 Web transfers are all elastic applications.
security
 sending host,
 a transport protocol can encrypt all data transmitted by
the sending process,
 receiving host,
 the transport-layer protocol can decrypt the data before
delivering the data to the receiving process
 other security services
 encryption,
 data integrity,
 end-point authentication …
TCP service: UDP service:
 reliable transport between sending and  unreliable data transfer between
receiving process sending and receiving process
 flow control: sender won’t overwhelm  does not provide: reliability, flow
receiver control, congestion control,
 congestion control: throttle sender timing, throughput guarantee,
when network overloaded security, or connection setup.
 does not provide: timing, minimum
throughput guarantee, security
 connection-oriented: setup required
between client and server processes