TCP/IP Model vs. OSI Model | Similarities and Differences | Fortinet (2024)

The Transmission Control Protocol/Internet Protocol (TCP/IP) model came before the Open Systems Interconnection (OSI) model, and it has five layers:

1. Application layer
2. Transport layer
3. Network access layer
4. Network interface layer
5. Hardware layer

This may look drastically different from the OSI model, primarily because some functions are encompassed in a single layer: the application layer. In TCP/IP, this provides users with the physical standards, transport functions, network interface, and internetworking functions that correspond with the first three layers of the OSI model. In other words, in the TCP/IP model, these services are all done in the application layer.

Here is a breakdown of what each layer in the TCP/IP model does:

1. Application layer. The application layer is where data originates on the sender’s side. Applications are used to create the data. A web browser, for example, is used to generate the data that gets sent through the rest of the layers, assisted by the Domain Name System (DNS), which associates web domain names with their Internet Protocol (IP) addresses.
2. Transport layer. In the transport layer, the data gets encoded so it can be transported through the internet using either the User Datagram Protocol (UDP) or TCP.
3. Network access layer. In the network access layer, the data gets a header and a trailer, and these tell the data where to go. This information is then conveyed to the network interface layer.
4. Network interface layer. At the network interface layer, the packet of data gets formatted and prepared to be transported and routed through the network.
5. Hardware layer. On the hardware layer, the data is turned into something that can be sent to and read by a computer or other device. For example, the IEEE 802.3 protocol is used to convert data into what is used in an Ethernet connection.

The OSI model is another way of transmitting data over the internet. The biggest difference between the OSI and TCP/IP models is the OSI model has seven layers instead of five. Although both the TCP/IP and OSI models transport data, the ways they send it are slightly different, which is sometimes why TCP/IP is used instead of OSI.

However, in the TCP/IP vs. OSI discussion, there are more similarities between the OSI and TCP/IP models than differences. Both provide data communication services, enabling users to send and receive information from their IP address using the services made available by their internet service provider (ISP).

The OSI model has the following layers:

1. Physical. This consists of a data connection between a device generating data and the network.
2. Datalink. The datalink layer is the point-to-point connection that transmits the data to the network layer.
3. Network. In the network layer, the data gets its address and routing instructions in preparation for its journey across the network.
4. Transport. In the transport layer, the data hops between different points on the network on its way to its destination.
5. Session. The session layer has a connection that manages the sessions happening between applications.
6. Presentation. The presentation layer is where data gets encrypted and decrypted and converted into a form that is accessible by the application layer,
7. Application. In the application layer, an application, such as an internet browser, gets the data and a user can then interact with it.

Both the TCP/IP and OSI models provide logical ways of networking, as well as the processing of information using a layered system. In both systems, each layer has a specific function. This makes it easier to pinpoint where issues are occurring in the event of a failure.

For example, in both models, you can determine if the data is not being transmitted properly to a hardware device by isolating potential problems in the data link layer (OSI) or the hardware layer (TCP/IP).

The biggest difference between the two models is that the OSI model segments multiple functions that the TCP/IP model groups into single layers. This is true of both the application and network access layers of the TCP/IP model, which contain multiple layers outlined within the OSI model.

This is a significant difference because it can make it more difficult to troubleshoot issues or enhance performance when you are using the TCP/IP model. With the OSI model, for example, you can focus specifically on the application layer, presentation layer, or session layer to figure out why data is not coming out the way you expect.

With the TCP/IP model, on the other hand, the functions of these three layers are all combined in the application layer. In other words, while someone may say, “There is a problem in the application layer,” someone thinking in terms of the OSI model may get confused, wondering, “OK, but which part of the application layer?”

For many users, the OSI model is the preferable choice—all other things being equal. The fact that it segments the functions of the network into more layers makes troubleshooting and improving network performance more straightforward.

However, TCP/IP has the advantage of having more applications, and it is also commonly used in more current networking structures. This makes it preferable for some administrators because it may be the framework commonly used by teammates or other admins.