The seven layers of the OSI reference model can be divided into two categories: upper layers and lower layers. The upper layers are typically concerned only with applications, and the lower layers primarily handle data transportation. The sections that follow examine the three upper layers, the four lower layers, and the functions of each.
The upper layers of the OSI reference model—5, 6, and 7—are concerned with application issues. They are generally implemented only in software programs. The application layer is the highest layer and is closest to the end user. Both users and application layer processes interact with software programs that contain a communications component so that the application can interact with the OSI model effectively. The sections that follow review the functions of each upper layer in detail.
NOTE The term upper layer is often used to refer to any higher layer, relative to a given layer. The opposite, lower layer, is used to refer to any layer below the one being discussed.
The application layer essentially acts as the end-user interface. This is the layer where interaction between the mail application (cc:Mail, MS Outlook, and so on) or communications package (Secure CRT for Telnet or FTP Voyager for FTP) and the user occurs. For example,
10 Chapter 1: Networking and Routing Fundamentals when a user wants to send an e-mail message or access a file on the server, this is where the process starts. Another example of the processes that occur at this layer are network file system (NFS) use and the mapping of drives through Windows NT.
The presentation layer is responsible for the agreement and translation of the communication format (syntax) between applications. For example, the presentation layer enables Microsoft Exchange to correctly interpret a message from Lotus Notes. A historical example of why the presentation layer is needed is when a sender is transmitting in EBCDIC (8-bit) character representation to a receiver that needs ASCII (7-bit) character representation. Another example of the actions that occur in this layer is the encryption and decryption of data in Pretty Good Privacy (PGP).
The session layer responsibilities range from managing the application layer's transfer of information to the data transport portion of the OSI reference model. An example is Sun's or Novell's Remote Procedure Call (RPC), which uses Layer 5.
The lower layers of the OSI reference model—1, 2, 3, and 4—handle data transport issues. The physical and data link layers are implemented in hardware and software. The other lower layers are generally implemented only in software. These lower layers are the ones that network engineers and designers need to focus on to be successful. The sections that follow review the functions of each of the lower layers in detail.
The transport layer is responsible for the logical transport mechanism, which includes functions conforming to the mechanism's characteristics. For example, the transmission control protocol (TCP), a logical transport mechanism, provides a level of error checking and reliability (through sequence numbers) to the transmission of user data to the lower layers of the OSI reference model. This is the only layer that provides true source-to-destination, end-to-end connectivity through the use of routing protocols such as open shortest path first (OSPF) or the file transfer protocol (FTP) application as examples of TCP.
Contrast the presence of TCP with the user datagram protocol (UDP), which is an unreliable protocol that does not have the additional overhead that provides error checking and reliability like TCP. Some common examples of UDP-based protocols are Trivial File Transfer Protocol (TFTP) and Simple Network Management Protocol (SNMP). The most common usage of UDP is streaming media solutions, such as Real Audio.
Understanding the Seven Layers of the OSI Reference Model 11
The network layer determines a logical interface address. Routing decisions are made based on the locations of the Internet protocol (IP) address in question. For example, IP addresses establish separate logical topologies, known as subnets. Applying this definition to a LAN workstation environment, the workstation determines the location of a particular IP address and where its associated subnet resides through the network layer. For example, there might be subnet 10.10.10.x, where the customer service people have their workstations or servers, and another subnet 10.20.20.x, where the finance people have their servers or workstations. IP addressing is discussed in more detail later in the section "Internet Protocol Addressing." Until then, remember that a logical IP address can have three components: network, subnet, and host.
Layer 2—Data Link
The data link layer provides framing, error, and flow control across the network media being used. An important characteristic of this layer is that the information that is applied to it is used by devices to determine if the packet needs to be acted upon by this layer (that is, proceed to Layer 3 or discard). The data link layer also assigns a media access control (MAC) address to every LAN interface on a device. For example, on an Ethernet LAN segment, all packets are broadcast and received by every device on the segment. Only the device whose MAC address is contained within this layer's frame acts upon the packet; all others do not.
It is important to note at this point that serial interfaces do not normally require unique Layer 2 station addresses, such as MAC addresses, unless it is necessary to identify the receiving end in a multipoint network. On networks that do not conform to the IEEE 802 standards but do conform to the OSI reference model, the node address is called the data link control (DLC) address. For example, in Frame Relay, this Layer 2 address is known as the data-link connection identifier (DLCI).
MAC addresses are 6 bytes or 48 bits in size, of which 24 bits are dedicated for Organization Unique Identification (OUI) and 24 bits are for unique identification. See the Institute of Electrical and Electronic Engineers (IEEE) website for more information.
The IEEE assigns Ethernet address blocks to manufacturers of Ethernet network interface cards. The first 3 bytes of an Ethernet address are the company ID, and the last 3 bytes are assigned by the manufacturer. Table 1-2 shows an example of an Ethernet address that is assigned to Cisco Systems.
12 Chapter 1: Networking and Routing Fundamentals
Table 1-2 Example Ethernet Address
Organization Unique ID Assigned by Cisco
When discussing MAC addresses, some people refer to the Organization Unique IDs as the vendor ID or OID. All are correct; however, the IEEE uses the term shown in Table 1-2.
Layer 1 —Physical
The physical layer, the lowest layer of the OSI reference model, is closest to the physical network medium (for example, the network cabling that connects various pieces of network equipment). This layer is responsible for defining information regarding the physical media, such as electrical, mechanical, and functional specifications to connect two systems. The physical layer is composed of three main areas: wires, connectors, and encoding. Figure 1-3 shows the relationship among the seven layers.
Figure 1-3 Detailed OSI Layer Relationships
Name of Unit Exchanged
Figure 1-3 Detailed OSI Layer Relationships
Name of Unit Exchanged
OSI Reference Model Layers and Information Exchange 13
Was this article helpful?