Communication Networks/OSI Reference Model
This page will discuss the OSI Reference Model
Contents |
[edit] OSI Model
| Layer | What It Does |
|---|---|
| Application Layer | The application layer is what the user of the computer will see and interact with. This layer is the "Application" that the programmer develops. |
| Presentation Layer | The Presentation Layer is involved in formatting the data into a human-readable format, and translating different languages, etc... |
| Session Layer | The Session Layer will maintain different connections, in case a single application wants to connect to multiple remote sites (or form multiple connections with a single remote site). |
| Transport Layer | The Transport Layer will handle data transmissions, and will differentiate between Connection-Oriented transmissions (TCP) and connectionless transmissions (UDP) |
| Network Layer | The Network Layer allows different machines to address each other logically, and allows for reliable data transmission between computers (IP) |
| Data-Link Layer | The Data-Link Layer is the layer that determines how data is sent through the physical channel. Examples of Data-Link protocols are "Ethernet" and "PPP". |
| Physical Layer | The Physical Layer consists of the physical wires, or the antennas that comprise the physical hardware of the transmission system. Physical layer entities include WiFi transmissions, and 100BaseT cables. |
[edit] What It Does
The OSI model allows for different developers to make products and software to interface with other products, without having to worry about how the layers below are implemented. Each layer has a specified interface with layers above and below it, so everybody can work on different areas without worrying about compatibility.
[edit] Packets
Higher level layers handle the data first, so higher level protocols will touch packets in a descending order. Let's say we have a terminal system that uses TCP protocol in the transport layer, IP in the network layer, and Ethernet in the Data Link layer. This is how the packet would get created:
1. Our application creates a data packet
- |Data|
2. TCP creates a TCP Packet:
- |TCP Header|Data|
3. IP creates an IP packet:
- |IP Header|TCP Header|Data|CRC|
4. Ethernet Creates an Ethernet Frame:
- |Ethernet Header|IP Header|TCP Header|Data|CRC|
On the receiving end, the layers receive the data in the reverse order:
1. Ethernet Layer reads and removes Ethernet Header:
- |IP Header|TCP Header|Data|CRC|
2. IP layer reads the IP header and checks the CRC for errors
- |TCP Header|Data|
3. TCP Layer reads TCP header
- |Data|
4. Application reads data.
It is important to note that multiple TCP packets can be squeezed into a single IP packet, and multiple IP packets can be put together into an Ethernet Frame.