The OSI model is the ideal infrastructure for the computer networks and communication. In the real world scenario, the OSI model is a standard but was never implemented and TCP/IP model is widely accepted and used. Decades back when the researchers were trying to build computer networks they had two options namely OSI model and TCP/IP model. In those times the OSI model didn’t seemed as an advantage over the TCP since it had too many addresses (hexadecimal addressing) and hence the TCP/IP suite of protocols were elected over OSI suite of protocols. Later, as years went on, the Internet became popular and hence it started growing and was about to hit a global crisis of IP addresses.
The people realized that actually OSI model was a better option over TCP/IP but until then it was too late. There were 3 reasons why OSI didn’t replace TCP. Firstly, the OSI model was completed after TCP/IP was completely in place and a lot of money had been already spent on the suite, hence, changing it would cost a lot. Secondly, some layers in the OSI model such as the presentation layer and session layer were never fully defined. That is, the protocols used in those mentioned layers was never fully defined, described nor developed. Thirdly, when the OSI model was practically implemented by a private network with in an organization it didn’t show enough efficiency as compared to TCP/IP model and hence the Internet was not switched from TCP/IP to OSI model. Therefore, to deal with the rapid rising of IP address crises an advanced version of addressing has been developed which was similar to OSI model and was known as IPv6.
If you remember we have discussed briefly about IPv6 in one of the previous article “Noob’s Guide Series : IP address”. If you have read that article then I would recommend you to read it before we dive into the concepts.
Now let’s move further with our discussion on OSI model. The OSI model has 7 layers which are associated with different tasks which enable communication over the computer networks and are listed below:
1. Physical layer
2. Data link layer
3. Network layer
4. Transport layer
5. Session layer
6. Presentation layer
7. Application layer
TCP/IP model consists of 5 layers namely, application layer, transport layer, network layer, data link layer and physical layer.
Physical layer: The physical layer is the lowest layer in both the models. The task of this layer is to carry bits in a packet (also known as a frame in data-link layer) across the link from one computer to another. This layer is basically acts a transmission medium for data across the network in the form of electromagnetic signals.
Data-link layer: This layer particularly is responsible for transportation of the datagram across the link. TCP/IP suite does not define any specific protocol for data-link layer. It supports all the standard protocols. The data-link layer takes the datagram and encapsulates it in a packet called frame.
Network layer: This layer has the major role in both the models. It is responsible for communication between the host & destination and it is also responsible for proper routing of the packets across the network. The Internet Protocol (IP) is included in the network layer which describes the packet format, structure of address and is also responsible for routing of a packet from a source to destination. The network layer also has some auxiliary protocols which further help IP in its routing tasks.
Transport layer: When a user does some activity in the application layer the message is sent to the transport layer which encapsulated into a transport layer packet called a user datagram. The transport layer has two major protocols. First one is the Transmission Control Protocol (TCP) and the second one is the User Datagram Protocol (UDP). They are used based on the request from the application layer. The TCP protocol is used when we need a connection oriented communication where is the UDP is used when we need a connection-less communication. What TCP does is, it creates a logical pipe between the hosts and the data is transmitted across the that logical pipe for that particular session. Though TCP is quite slow as compared to UDP, it reduces the chances of data loss, provides flow control and error control. The UDP can be used when we need to send a considerable amount of information. It doesn’t provide flow control, error control and congestion control.
Application layer: Basically, when we think of communications in computer networks, we would notice that the whole communication looks like it is happening on the application layer, but the fact is, all the other layers are involved during the transmission of the packets. The application layer communicates to the transport layer which in turn communicates to the network layer and which is further carried to data-link layer and finally to physical layer where the transmission takes place. The destination host receives the packets at its physical layer and the packet is taken till the application layer after processing at each stage/layer. The application layer includes a bunch of protocols which are designed to enable different types of communications, such as HTTP for web, SMTP for email, SSH for secured shell which is used for remotely accessing a system, etc. The last 3 layers of OSI model (i.e., session layer, presentation layer and application layer) are integrated into a single application layer in TCP/IP model.
Comments
Post a Comment