Client-Server Interaction

1
Client-Server Interaction

• Based on Appendix 1 in Computer Networks and
  Internets, Comer

2
Network Applications

• All of the lower layers of the network
  communication protocol stack exist so that
  applications on different computers can interact.
  
• Applications are high-level software used
  directly by users.
• The applications (e.g. Excel) sit on top of the
  application layer which provides various
  network-related services (e.g. FTP) to the
  applications.
• The basic paradigm for this application-to-applica
  tion interaction on an internet is the
  client-server model.

3
Application Software

• Applications and the application layer hide all
  of the details of the network from the user.
• Information hiding
• In most cases the user does not have to know the
  IP address (let alone the MAC address) of the
  computer he or she is communicating with.
• The application provides a user-friendly symbolic
  name, which is then translated into the
  corresponding number.
• E.g. a domain name service

4
Connection-Oriented

• While TCP is in part responsible for establishing
  a session between two computers, the TCP does not
  establish a connection on its own volition.
• The source application layer must direct its
  transport layer to do so.
• The destination application layer must direct its
  transport layer to listen for others trying to
  connect to it.
• In the phone analogy, the source must dial the
  number, the destination must hear the phone ring
  and answer it.

5
The client-server paradigm

• The approach in which the source initiates
  contact with a listening destination is said to
  adhere to the client-server paradigm.
• The initiating source is called the client.
• Clients are active.
• The listening destination is called the server.
• Servers are passive.

6
Two meanings
2

• In one sense, client-server is used to refer to
  the communication model
• Client Active requester
• Server Passive listener/responder
• In another sense, client-server is used to refer
  to machines which are designed to act as clients
  or as (dedicated) servers
• In this sense, a server is a machine capable of
  offering a such service to a number of clients.

7
What is a Server?

• A server can refer to a program that passively
  waits for communication.
• A server can refer to a computer dedicated to
  running server programs.
• Sometimes the term server class computer is
  used for the latter case.
• But most of the world uses the term server in
  both cases.
• Servers or server-class computers are typically
  powerful machines which run server software.

8
Not the only paradigm in town

• Client-server is not the only model for
  computer-to-computer communication.
• Peer-to-peer A type of network in which each
  workstation has equivalent capabilities and
  responsibilities. Peer-to-peer networks are
  generally simpler, but they usually do not offer
  the same performance under heavy loads.
• Master/slave master polls the slaves to see if
  they have anything to transmit.

9
Mainframe architecture

• Mainframes are an example of the master-slave
  model.
• Most of the data storage and computing power lies
  in the mainframe (a central computer).
• Users interact with the mainframe via terminals
  (dumb terminals) which is little more than a
  monitor and keyboard.
• In the early days of PCs, they were often
  disabled to work as dumb terminals in a
  mainframe.

10
Thin and fat clients

• There was an idea of returning to some extent to
  the notion of a dumb terminal.
• In client/server applications, a client designed
  to be especially small so that the bulk of the
  data processing occurs on the server is called a
  thin client.
• A client that performs the bulk of the data
  processing operations is called a fat client.
• Although the term usually refers to software, it
  can also apply to a relative abilities of a
  network computer.

11
Client/Server Operating Systems

• Often the version of the operating system one
  loads on a computer that mainly plays the client
  role will be different from the version of the
  operating system loaded on the server
• E.g. Windows 7 is different from Windows 2008
  Server

12
The client

• Client software
• Requests service but also performs its own
  computations.
• Is invoked directly by a user and executes only
  for one session
• as opposed to starting automatically and running
  continuously
• Usually runs locally on a users PC
• Initiates contact with a server.
• May access multiple services, but only
  communicates with one server at a time.
• Does not require special hardware (beyond
  standard communications hardware) or Operating
  System.

13
The Server

• Server software
• Is a special-purpose, privileged program
  dedicated to providing one service.
• Can handle multiple remote clients at the same
  time.
• Is invoked automatically (typically when a system
  boots) and continues to execute through many
  sessions.
• Waits passively for contact from arbitrary remote
  clients.
• Accepts contact from arbitrary clients, but
  offers a single service.
• Requires powerful hardware and a sophisticated
  operating systems.

14
Application Software

• Applications (as opposed to services provided by
  the application layer) that communicate with
  other applications are often written with either
  the client role or server role specifically in
  mind.
• Thus one talks about client-side scripting and
  server-side scripting.

15
Scripting

• Client-side scripting is code written for the
  client end of a client-server system.
• For example, JavaScript scripts are client-side
  because they are executed by your browser (the
  client).
• Server-side scripting is code written for the
  server end of a client-server system.
• For example, PHP scripts are server-side because
  they run on the Web server.
• Java applets can be either server-side or
  client-side depending on which computer (the
  server or the client) executes them.

16
Client-Server Interaction

• Information between client-server passes in both
  directions.
• Clients request, servers respond.
• Sometimes the servers response is an ongoing
  communication based on one initial request.
• Client-server applications lead the protocol
  suite to establish the session and send/receive
  information.
• A computer needs all stacks of the software
  protocol suite to run as a client or server.

17
Client-Server
18
One connection/Many services

• A computer with multiple servers still requires
  only a single physical connection for
  communication.
• Different services will be associated with
  different ports.
• Since servers are often in passive modes, having
  one computer offer many services can help reduce
  administrative overhead and cost without a
  significant reduction in performance.
• Virtualization
• However, if many clients are expected for a
  particular service it may be worthwhile having a
  machine dedicated to it.

19
Some Server Types

• A file server is a computer and storage device
  dedicated to storing files. Any user on the
  network can store files on the server.
• A print server is a computer that manages one or
  more printers.
• A network server is a computer that manages
  network traffic.
• A database server is a computer system that
  processes database queries.

20
Multiple services
21
A thread for each request

• In order to handle concurrency, that is, dealing
  with many clients at once, the server-class
  computer uses threads.
• Certain services being associated with well-known
  ports.
• Actually the main thread or listener (usually a
  daemon) is located at this port, but as soon as a
  request is made, a new thread is spun off
  (forked) for it.
• This allows the main listening thread to await
  the next client.

22
Thread

• A thread is the information needed to serve one
  individual user or a particular service request.
• If multiple users are using the program or
  concurrent requests from other programs occur, a
  thread is created and maintained for each of
  them.

23
Types of Transport Protocols

• The application services may be
  connection-oriented or connectionless.
• Connection-Oriented
• The applications must first establish the
  connection and then send the data across the
  connection. TCP requests the connection and once
  established the communication begins. Session
  must be terminated.

24
Types of Transport Protocols

• Connectionless Interface
• The application sends a message to the
  destination at any time. The sending application
  needs to specify the destination with each
  message sends. The UDP is the connectionless
  support mechanism in the TCP/IP protocol.
• For a client request using the connectionless
  interface, servers often require the request to
  be sent in a single message. The server responds
  in a single message.

25
Service Connections Choice

• Servers may provide a service through both
  connection methods.
• The choice of transport is dependent on the
  client.
• This choice requires the server to provide both
  transport implementations. This may be
  implemented by
• Setting up 2 servers for the same service, one
  for connectionless transport and one for
  connection-oriented transport.
• Setting up a single server that is able to
  interact with either type of transport protocol
  simultaneously.

26
Client-Server Interactions

• A client application is not restricted to
  accessing a single service. A single application
  can become the client of several servers over
  time.
• A client application is not restricted to
  accessing a single server for a given service. A
  client might send a request to multiple servers
  to improve performance.
• A server is not restricted from performing
  further client-server interactions . A server
  for one service can become a client of another.

27
Two-tier

• The term two-tier refers to client/server
  architectures in which the user interface runs on
  the client and the database is stored on the
  server.
• The actual application logic can run on either
  the client or the server.

28
Three tier

• A special type of client/server architecture
  consisting of three well-defined and separate
  processes, each running on a different platform
• The user interface, which runs on the user's
  computer (the client).
• The functional modules (business rules) that
  actually process data. This middle tier runs on a
  server and is often called the application
  server.
• A database management system (DBMS) that stores
  the data required by the middle tier. This tier
  runs on a second server called the database
  server.

29
Three tier (Cont.)

• The three-tier design has many advantages over
  traditional two-tier or single-tier designs,
  mainly
• The added modularity makes it easier to modify or
  replace one tier without affecting the other
  tiers. (Remember Layering and information
  hiding.)
• Separating the application functions from the
  database functions makes it easier to implement
  load balancing, i.e. spreading the processing out
  

30
References

• Computer Networks and Internets, Comer
• http//www.whatis.com
• http//www.webopedia.com