How does the OSI model work?

The OSI model separates network communication into seven layers that help engineers understand how data moves between connected systems.

Key Highlights

  • The OSI model divides network communication into seven distinct layers, each responsible for specific functions in data transmission.
  • Upper layers (application, presentation, session) handle data formatting, encryption, and session management, facilitating user interactions.
  • Lower layers (transport, network, data link, physical) manage data transfer, addressing, routing, and physical transmission media.

NASHUA, N.H. — Sending data from one networked system to another involves several steps. The data must be prepared for transmission, directed toward the correct destination, carried across a connection and delivered in a form the receiving system can use.

The Open Systems Interconnection (OSI) model provides a framework for understanding where those different functions take place. It divides network communication into seven layers- physical, data link, network, transport, session, presentation and application.

The International Organization for Standardization published the model in 1984. Modern internet communications generally rely on the TCP/IP model instead, but engineers still use the OSI model to describe network functions and isolate communications problems.

Related: What is the OSI model and what are its seven layers?

How does data move through the OSI model?

The seven layers describe different parts of the communication process, beginning with the applications that use network services and ending with the physical transmission of data. At the top is the application layer, where software accesses services needed to communicate over a network. Protocols associated with web traffic, email, file transfers and domain-name lookups operate at this level.

The presentation layer deals with how information is represented before another system receives it. It can handle tasks such as encryption and compression and ensure that the receiving system can interpret the data.

The session layer deals with the exchange between two communicating systems. It helps organize that exchange over time, including opening and closing sessions and providing points from which a transfer can resume if communication is interrupted. Together, those upper layers deal largely with the information being exchanged and the interaction between the systems using it.

The transport layer moves the process closer to delivering that information from one host to another. Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) operate at this layer, although they take different approaches to transporting data. TCP provides mechanisms for reliable delivery, while UDP sends data without the same built-in reliability mechanisms.

How does data reach the correct destination?

The network layer handles logical addressing and routing between networks. Internet Protocol (IP) operates here, with packets carrying source and destination IP addresses as they move across networks.

Below that, the data link layer handles communication across a local network link. It organizes data for transfer between devices on that link and includes functions associated with Media Access Control (MAC) addresses.

The physical layer sits at the bottom of the model. At this point, communication depends on the equipment and medium carrying the information. The layer covers the electrical, optical or radio characteristics used to transmit raw bits across a connection.

The process reverses when the information reaches the receiving system. Functions associated with the lower layers handle the incoming transmission before the data moves upward toward the application that will use it.

Why do engineers use the OSI model for troubleshooting?

Separating network communication into layers gives engineers a way to narrow down where a problem is occurring. A physical connection could be unavailable while the higher-level network configuration remains correct. In another case, the physical link could work normally while a problem with addressing or routing prevents traffic from reaching its destination.

The OSI model also gives engineers common terminology for distinguishing between network functions. The framework can help engineers describe how equipment and protocols from different vendors interact. Because the model defines functions independently of a particular manufacturer or technology, it provides a common reference point when discussing network behavior.

How is the OSI model different from TCP/IP?

The OSI and TCP/IP models organize some of the same networking functions differently.

OSI uses seven layers. TCP/IP uses fewer and combines functions that OSI separates. The application, presentation and session functions described by the upper three OSI layers generally fall within the TCP/IP application layer. Functions associated with the OSI physical and data link layers are grouped within the TCP/IP network-access layer.

TCP/IP forms the basis of modern internet communications, while OSI serves primarily as a reference model for understanding and describing networking functions.

About the Author

Samantha McGrail

Associate Editor

Sign up for our eNewsletters
Get the latest news and updates

Voice Your Opinion!

To join the conversation, and become an exclusive member of Military Aerospace, create an account today!