Skip to main content

TCP/IP Model


TCP-IP Model
Layers of TCP-IP Reference Model
Application
Transport
Network/Internet Protocol
Host to network


Host to network Interface Layer:
  • This layer defines the protocols & hardware request to deliver data across physical network. 
  • The term network interface layer refers to the fact that it defines how to connect the host computer which is not a part of the network, to the network.
Ethernet is one example protocol at TCP/IP network layer. TCP/IP reference model does not really says much about what happens where, expect to point out that the host has to connect to the network using protocols so it can send up packets over it.

Network/Internet Layer:
  • This layer defines an official packet format and protocols called Internet Protocol(IP). 
  • The job of this layer is to deliver IP packets where they are supposed to go. Packet routing is clearly the major issue at this layer and avoiding congestion. 
  • For these reasons, it is reasonable to say that TCP/IP internet layer is very similar in functionality to OSI network layer. 
  • This layer holds the whole architecture together. This job is to permit host to inject packets into any network. This may even arrive in a different order than they were sent, in this case, it is the job of higher layer to rearrange them, if an order delivery is required. 
  • IP defines logical address called IP address that allows each TCP/IP speaking devices called IP host to communicate. It also defines routing that is the process of how a router should forward or route the packets of data. For eg: Ordinary mail system
  Transport Layer:
  • TCP is a connection oriented protocol. Transport layer is designed to allow a wide stream originating on one machine to be delivered without error on another machine connected to internet. 
  • The 2 end to end protocols have been defined here:
  1. TCP – reliable connection oriented protocol
  2. UDP – unreliable connectionless protocol
TCP
It is reliable connection oriented protocol. It fragments the incoming wide stream into discrete messages and pass each one to internet layer. At the destination, the receiving TCP process reassembles the received messages into output stream. TCP also handles flow control to make sure a fast sender cannot flood a slow receiver with more messages that it can handle.
UDP
The second protocol of this layer is UDP. It is an unreliable connectionless protocol that do not want TCP sequencing and flow control and wish to define their own. It is also widely used for one short client server type request reply queries and application in which fast delivery is more important than accurate delivery such as transmitting speech. UDP is a communication protocol that does not perform validity check on data. It is used in video conferencing.

Difference between TCP and UDP
TCP
UDP
Connection oriented
Connectionless
Reliability in delivery of messages
No recovery is performed for any lost message
Splitting messages into datagram
No reassembly and synchronization
Keeps track of sequence
No sequencing and in case of error, the whole message is retransmitted
Uses checksum for detecting errors and windows for acknowledgement
No windows and acknowledgement mechanism
Reliability is a must
Server and client messages fit completely within a packet
Overhead low but higher than UDP
Overhead is very low
Speed is high but not as high as of UDP
Speed is very high
Eg: FTP, SMTP, HTTP
Eg: SNMP, TFTP, DHCP

Application Layer:
  • NNTP(Network News Transfer Protocol) is used for moving news articles around and http is used for patching pages on www.
  •  This layer contains all the higher layer protocols. TCP/IP does not have a session and presentation layer. 
  • TCP/IP model have higher level protocols such as TELNET. It allows users on one machine to login to a distant machine and work from there. 
    • TCP/IP Model


  • FTP provides a way to move data efficiently from one machine to another. 


  • DNS(Domain Name Server/System/Service) for mapping host names onto their network address. 
Labels:

Comments

Post a Comment

Popular posts from this blog

Advantages and Disadvantages of EIS Advantages of EIS Easy for upper-level executives to use, extensive computer experience is not required in operations Provides timely delivery of company summary information Information that is provided is better understood Filters data for management Improves to tracking information Offers efficiency to decision makers Disadvantages of EIS System dependent Limited functionality, by design Information overload for some managers Benefits hard to quantify High implementation costs System may become slow, large, and hard to manage Need good internal processes for data management May lead to less reliable and less secure data
1 comment
Read more

Inter-Organizational Value Chain

The value chain of   a company is part of over all value chain. The over all competitive advantage of an organization is not just dependent on the quality and efficiency of the company and quality of products but also upon the that of its suppliers and wholesalers and retailers it may use. The analysis of overall supply chain is called the value system. Different parts of the value chain 1.  Supplier     2.  Firm       3.   Channel 4 .   Buyer
Post a Comment
Read more

Big-M Method and Two-Phase Method

Big-M Method The Big-M method of handling instances with artificial  variables is the “commonsense approach”. Essentially, the notion is to make the artificial variables, through their coefficients in the objective function, so costly or unprofitable that any feasible solution to the real problem would be preferred, unless the original instance possessed no feasible solutions at all. But this means that we need to assign, in the objective function, coefficients to the artificial variables that are either very small (maximization problem) or very large (minimization problem); whatever this value,let us call it Big M . In fact, this notion is an old trick in optimization in general; we  simply associate a penalty value with variables that we do not want to be part of an ultimate solution(unless such an outcome is unavoidable). Indeed, the penalty is so costly that unless any of the  respective variables' inclusion is warranted algorithmically, such variables will never be p
Post a Comment
Read more