Interconnection Models

The goal of every interconnection model is to provide standards to which all computer hardware and software vendors can adhere so that the present multiplicity of interconnection and interface practices is reduced, thus reducing the costs of designing and producing both hardware and software. The existing models are covered in literally hundreds of sources in great detail, but the following section provides a short description of the DoD model and the seven-layer OSI model from a standpoint of understanding the troubleshooting process.

Department of Defense Model

The efforts to create an interface model for internetworking started with the four-layer Department of Defense (DoD) model, which was developed by the Department of Defense in the 1970s for the Defense Advanced Research Projects Agency (DARPA) Internetwork Project (see Figure 4-1). The core Internet protocols adhere to this model, although the OSI model is preferred for new designs.

Figure 4-1. The Four Layers of the Department of Defense (DoD) Model

The four layers of the DoD model, from bottom to top, are as follows :

• Network access layer Responsible for delivering data over the hardware media. Different protocols are selected from this layer, depending on the type of physical network.

• Internet layer Responsible for delivering data across a series of different physical networks that interconnect a source and a destination machine. Routing protocols are most closely associated with this layer, as is the Internet Protocol (IP), which is the Internet's fundamental protocol.

• Host-to-host layer Handles connection rendezvous, flow control, retransmission of lost data, and other generic data flow management. The mutually exclusive Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) are this layer's most important members .

• Process layer Contains protocols that implement user-level functions, such as mail delivery, file transfer, and remote login.

Seven-Layer OSI Model

In the 1980s, the European-dominated International Standards Organization (ISO) began to develop its OSI networking suite. OSI has two major components : an abstract model of networking (the basic reference model, or seven-layer model ), and a set of concrete protocols. This model eventually grew into the Internet, to which the core Internet protocols adhere. This section discusses the layers of the OSI model.

The physical layer is concerned with transmitting raw bits over a communication channel. The design issues are related to timing, line coding, and framing. Categories of the physical layer include simplex, half duplex, and full duplex. The layer deals with the physical (electrical and mechanical) aspects of transmitting data (such as voltage levels, pin-connector design, cable lengths, and grounding arrangements).

The main activity of the data link layer is to deal with the transmission of data frames over a physical link between network entities, including the incorporation of error-correcting coding into the data frames. It accomplishes this task by having the sender fragment the input data into data frames (typically a few hundred bytes), transmit the frames sequentially, and process the acknowledgment frames that are sent back by the receiver. It is up to the data link layer to create and recognize frame boundaries that can be accomplished by attaching special bit patterns to the beginning and end of the frame. If there is a chance that these bit patterns might occur within the data, special care is taken to avoid confusion. This layer is responsible for error control between adjacent nodes, and consequently, flow regulation and error handling are integrated.

The network layer controls the operation of the subnet. The network layer establishes paths for data between a pair of computers, handles any switching among alternative routes between the computers, and also contains definitions of how to break the data segments up into individual packets of data. In this way, packets can be transmitted and reassembled. A key design issue is determining how packets are routed from source to destination. Routes can be static, when the routing does not depend on network conditions, and dynamic, when routes can be dynamically determined based on routing protocol decisions that reflect network conditions. An excessive number of packets in the subnet require sophisticated congestion control, which is another function of the network layer. When a packet has to travel from one network to another to get to its destination, many problems can arise. The addressing used by the second network might be different from the first one. The second one might not accept the packet at all because it is too large, or the protocols might differ , and so on. It is up to the network layer to overcome all routing problems to allow heterogeneous networks to be interconnected . In broadcast designs, the routing problems are simple, and as a result the functions of this layer are reduced.

The basic function of the transport layer is to accept data from the session layer, split it up into smaller units if necessary, pass these to the network layer, and ensure that the pieces (segments) all arrive correctly at the other end, performing the mandatory flow control. If the transport connection requires a high throughput, the transport layer might create multiple network connections by dividing the data among the network connections to improve throughput. The transport layer also determines what type of service to provide to the session layer, and ultimately, the users of the network. The most popular type of transport connection is an error-free point-to-point channel that delivers messages in the order in which they were sent, based on connection-oriented service. In terms of the TCP/IP suite, this function is performed by the TCP protocol. Other kinds of transport protocols transport isolated messages with no guarantee (best-effort type) about the order of delivery or the broadcasting of messages to multiple destinations. This type of connection is connectionless, and in terms of the TCP/IP suite, it is performed by the UDP. The transport header carries information on what message belongs to which process, thus providing multiplexing for the processes. Also, there is a mechanism to regulate information flow, so that a fast host cannot overrun a slow one. Flow control between hosts is distinct from flow control between switches, although similar principles apply to both.

The session layer enables users on different machines to establish sessions between them. A session allows a user to log into a remote time-sharing system or to transfer a file between two machines. One of the services of the session layer is dialogue control. Sessions can allow traffic to flow in both directions at the same time, or in only one direction at a time. If traffic can only go one way at a time, the session layer helps to keep track of whose turn it is. A related session service is token management. For some protocols, it is essential that both sides do not attempt the same operation at the same time. To manage these activities, the session layer provides tokens that can be exchanged. Another session service is synchronization, which inserts checkpoints into the data stream and ensures that after excessive data transfer, if the session has been interrupted , only the data after the last checkpoint is repeated. Another example here can be when one application process requests access to another applications process (e.g., Microsoft Word importing a chart from Excel).

The presentation layer performs functions to ensure consistent syntax and semantics of the transmitted information. It deals with syntactic representation of data, such as agreement on character code (ASCII, extensions to ASCII, Unicode); .doc, .txt, .gif, and .jpg formats; data-compression and data-encryption methods , and representations of graphics (such as files using the .pic or .bmp formats). The presentation layer is also concerned with other aspects of information representation. For example, data compression can reduce the number of bits to be transmitted, and cryptography is frequently required for privacy and authentication.

The application layer deals with data generation and contains a variety of protocols that are commonly needed, such as terminal emulation based on virtual terminal software. Another application layer function is file transfer because different file systems have different naming conventions and ways of representing text lines. Transferring a file between two different systems requires handling these and other incompatibilities.

Each layer of the OSI model receives data from the layer beneath it and provides service for the layer above it. The exchange of information between layers is based on protocol data units (PDUs) and the encapsulation of the user data occurs in every layer. Therefore, every layer adds extra information (headers and trailers ) to perform its functions.

In reality, data is passed from one layer down to the next lower layer at the sending computer, until the physical layer finally transmits it onto the network cable. As the data is passed down to a lower layer, it is encapsulated into a larger unit because additional layer information is added. At the receiving end, the message is passed upwards to the desired layer, and as it passes upwards through each layer, the encapsulation information is stripped off.

As an industry standard, OSI is a framework that every new product complies with because it is already accepted and widely adopted. A variety of other specifications, including IEEE (www.ieee.org), RFC (www.ietf.org/rfc.html), Asynchronous Transfer Mode (ATM) Forum (www.atmforum.org), DSL Forum (www.dslforum.org), and the Internet Engineering Task Force (www.ietf.org) create sublayers and define different model structures. It is important to remember the definition of the communication subsystem (see Figure 4-2) that includes up to the first three OSI layers, because it is the main focus of remote access troubleshooting.

Figure 4-2. The Open System Interconnection (OSI) Model, Including Some of the Related Protocols and the Communications Subsystem