CompTIA A_ Certification All-In-One Exam Guide, Seventh Edition - Michael Meyers [400]
How will each computer be identified? If two or more computers want to talk at the same time, how do you ensure that all conversations are understood?
What kind of wire? What gauge? How many wires in the cable? Which wires do which things? How long can the cable be? What type of connectors?
If more than one PC accesses the same file, how can they be prevented from destroying each other’s changes to that file?
How can access to data and peripherals be controlled?
Clearly, making a modern PC network entails a lot more than just stringing up some cable! Most commonly, you have a client machine, a PC that requests information or services. It needs a network interface card (NIC) that defines or labels the client on the network. A NIC also helps break files into smaller data units, called packets, to send across the network, and it helps reassemble the packets it receives into whole files. Second, you need some medium for delivering the packets between two or more PCs—most often this is a wire that can carry electrical pulses; sometimes it’s radio waves or other wireless methods. Third, your PC’s operating system has to be able to communicate with its own networking hardware and with other machines on the network. Finally, modern PC networks often employ a server machine that provides information or services. Figure 23-1 shows a typical network layout.
Figure 23-1 A typical network
This section of the chapter looks at the inventive ways network engineers found to handle the first two of the four issues. After a brief look at core technology, the chapter dives into four specific types of networks. You’ll dig into the software side of things later in the chapter.
Topology
If a bunch of computers connect together to make a network, some logic or order must influence the way they connect. Perhaps each computer connects to a single main line that snakes around the office. Each computer might have its own cable, with all of the cables coming together to a central point. Or maybe all of the cables from all of the computers connect to a main loop that moves data along a track, picking up and dropping off data like a circular subway line.
A network’s topology describes the way that computers connect to each other in that network. The most common network topologies are called bus, ring, star, and mesh. Figure 23-2 shows the four types: a bus topology, where all computers connect to the network via a main line called a bus cable; a ring topology, where all computers on the network attach to a central ring of cable; a star topology, where the computers on the network connect to a central wiring point (usually called a hub); and a mesh topology, where each computer has a dedicated line to every other computer—the mesh topology is mostly used in wireless networks. There are also hybrid topologies, such as star bus or star ring, that combine aspects of the other topologies to capitalize on their strengths and minimize their weaknesses. You’ll look at the most important hybrid topology, star bus, in a moment, but for now, make sure you know the four main topologies!
Figure 23-2 Clockwise from top left: bus, ring, mesh, and star topologies
If you’re looking at Figure 23-2 and thinking that a mesh topology looks amazingly resilient and robust, it is—at least on paper. Because every computer physically connects to every other computer on the network, even if half of the PCs crash, the network functions as well as ever (for the survivors). In a practical sense, however, implementing a true mesh topology network would be an expensive mess. For example, even for a tiny network with only 10 PCs, you would need 45