OSI reference model

OSI reference model

The Open System Interconnection (OSI) is a logical primary architectural model which is physically not present. It was created by the International Organization for Standardization (ISO). Earlier when networks were first introduced to the world, computers could only interact with those computers which have identical operating system. OSI model came into use so that computers with different manufacturers could interact with one another. It is truly a general model with horizontal approach. 

OSI model is also known as hierarchical/layered architecture model. It consists of 7 layers, they are:

• Application layer
• Presentation layer
• Session layer
• Transport layer
• Network layer
• Data link layer
• Physical layer

Advantage

1. Designing and development of component is easy.
2. It facilitates during troubleshooting.
3. Computers having different operating system can also connect.

There are some network devices that works on every layer such as: servers, gateways, web/application servers, etc,. Whereas network devices like routers work in the network layer, switches and bridges works at data link layer and hubs in the physical layer. Let us study each and every layer with their applications.

Image courtesy:Flickr

Application layer

Application layer carry out functions such as:

1. Transportation of files.
2. Emails.
3. Client/server process .
4. Activities related to network management.
5. Data integrity and error recovery.
6. Login and password verification.
7. Manipulates the information, redirects/re-transfers the files.

It is the layer where the actual communication takes place. It checks for the availability of communicating partner, resources required for connection and applications necessary for communication.

Presentation layer

As the name suggests, presentation layer presents the data. It performs various functions such as:

1. Presents the information in such a way that is easy to understand..
2. Works as translator when two communicating systems uses different forms and syntax.
3. Performs coding and conversions.
4. Compresses/decompresses data.
5. Encryption/decryption.

Session layer

Session layer creates, manages and terminates the applications. On the network, it separates data from different applications. Communication between host to client can be done by 3 different ways:

1.  Simplex: It is a one-way communication in which data is sent but the receiver cannot reply back.
2. Half-duplex: It is a two-way communication in which data message is passed only in one direction to avoid collision. 
3. Full-duplex: It is a two-way communication in which users from both the side can send and receive messages at the same time.

It performs the following functions:

1. Synchronizes and manages conversations.
2. Controls logging on/off, authentication and session management.
3. During transmission, data is synchronized and marked properly to avoid data loss.

Transport layer

Transport layer is used for reliable connection. It is represented as segments. It establishes a logical connection between the sending host and the receiving client. The network protocols TCP and UDP works on this layer.

TCP: TCP stands for transmission control protocol. It is used to establish communication between nodes and exchange data. It is a connectionless communication protocol, commonly used in all applications that require guaranteed delivery due to its reliability.

UDP: UDP stands for user data protocol. It is used as an alternative of TCP/IP. It does not divides data into packets and also doesn't ensure delivery in sequential order. It uses checksum capability to check data. 

Transport layer is also responsible for the following functions:

1. Decides whether the data transmission should be in parallel path or single pathe.
2. Performs multiplexing, splitting and segmenting of data.
3. Guarantees data transmission.
4. Breaks data groups into segments.

Network layer

Network layer performs the following functions:

1. It manages logical network devices.
2. Tracks the location of the device.
3. Breaks data groups into packets. 
4. Provides logical addressing for the path determination of routers.
5. Acts as network controller.
6. Determines the best path to move data.

Routers work in the network layer.

Data link layer

The data link layer provides the following:

1. Flow control.
2. Network topology.
3. Transmission of data.

It converts the data packets into frames, and performs the following functions:

1. Uses MAC (media access control) and LLC (logical link control)
2. Adds error detection bits to the data to be transmitted for error detection.
3. Combines bits into bytes and bytes into frames.
4. Information to be transmitted to the physical layer are error controlled and synchronized.

Switches and bridges work on this layer

Physical layer

Physical layer sends and receives bits which are in the numerical value of 0 or 1. It communicates directly with various communication media and also performs the following functions:

1. Activation, maintenance and deactivation of physical connection.
2. Defines voltage and data rates for transmission purpose.
3. Converts digital bits into electrical signals.
4. Decides whether the transmission is simplex. half-duplex or full-duplex.

Hubs work in the physical layer.