level layers (say 5, 6, and 7) might be very simple or possibly nonexistent. However, in a very complex data communications system, all seven software layers may be implemented. Although there is no requirement for any hardware or software vendor to adhere to the principles  set  forth  in  the  OSI  model,  there  is  a worldwide  trend in the computer industry toward acceptance and conformance to these standards. About now, you may be asking yourself, what are these seven software layers (shown in figure 1-8), and why all the need for protocols? Don’t all computers work  in  binary?  Do  they  not  all  have  operating systems? If a computer wants to communicate with another system, do you not simply connect them together using some type of cable? The answers to these questions   are   yes,   yes,   and   yes;   however,   the commonalities seem to stop there. Ideally,   if   the   hardware,   network   software, application software, and cabling were all supplied by the same manufacturer, we would have relatively few problems  to  contend  with  when  we  design  and implement a network. Everything would work together rather smoothly. However, a computer manufacturer’s architecture  can  make  it  difficult  to  interconnect h a r d w a r e    o f f e r e d    b y    o t h e r    c o m p e t i ng manufacturers/vendors.   The   protocols   used   by communications devices are also highly complex and are often completely different from one manufacturer to another. Then, there is the network software. Network software from one LAN vendor usually won’t work on a competitor’s   network,   nor   will   the   application programs. Even the cabling must be selected for a specific local-area network. We could go on and on explaining the many incompatibilities that exist within these different areas, but the good news is that many hardware and software manufacturers/vendors  provide  interfaces.  These various types of interfaces (bridges, gateways, routers, and so on) allow networks to be compatible with one another. At this point, we briefly talk about the seven software layers defined in the OSI model to give you some idea of what they are and why they are needed. To illustrate how the OSI model works, we are using the analogy of sending a letter using the U.S. postal system. Layer l—The physical layer  is concerned with the transmission of the unstructured raw bit stream over a   physical   meduim.   It   addresses   the   electrical, mechanical, and functional interface to the carrier. It is the physical layer that carries the signals for all the higher layers, as follows: Voltages  and  pulse  encoding  of  bits Media and media interface (cables, connectors, NIC, and so on) Line discipline (full- or half-duplex) Pin assignments In our mail analogy, the mail truck and the highway provide the services of the physical layer. Layer 2—The data link layer provides error-free transmission  of  information  over  the  physical  medium. This allows the next higher layer to assume virtually error-free transmission over the link. The data link layer is responsible for getting data packaged and onto the network cable. It manages the flow of the data bit stream into and out of each network node, as follows: . Creates and recognizes frame boundaries l Checks received messages for integrity . Manages channel access and flow control . Ensures correct sequence of transmitted data The data link layer detects, and when possible, corrects errors that occur in the physical layer without using the functions of the upper layers. It also provides flow-control techniques to ensure link-buffer capacity is not exceeded. In our analogy, the data link layer is concerned  with  sending  the  mail  trucks  onto  the highway and making sure they arrive safely. Layer 3—The network layer  decides  which physical  pathway  the  data  should  take,  based  on network conditions, priorities of service, and other factors. Software on the network interface card must build the data packet so the network layer can recognize and route the data to the correct destination address. It relieves the upper layers of the need to know anything about  the  data  transmission  and  switching  technologies used to connect the systems. It is responsible for establishing,  maintaining,  and  terminating  connections across the intervening communications facility, as follows: Addresses messages Sets up the path between communicating nodes on possibly different networks Routes messages among networks Is concerned with the sequence delivery of data packets 1-14