LAN Topology Types

In the next three sections I will explain the three different network topologies you need to know for the exam. A network topology is all the combined physical, logical, or virtual components that make up the network or network segment. Although there are many topologies, we will focus on those necessary for the exam, which include the bus, ring, and star topologies.

Bus

A bus topology is a local area network (LAN) where each of the networked devices are attached to a single cable or link, as shown in Figure 3.1. In a bus topology, stations are attached to a linear multiport medium where only half-duplex operations exist between a station and a bus.

Figure 3.1. A bus topology LAN. Notice that all the workstations are connected by a single cable.

Half-duplex is where communication occurs bi-directionally on one cable. This means that a device sending data cannot receive data at the same time. Full-duplex uses two individual cables, one to send and another to receive. This allows a device to send at the same time it receives data.

Frames that are transmitted to the bus provide the address of the frame's destination. If the frame gets to the end of the link and the frame has not found its intended destination, then the frame is lost. In a bus topology, there is no security; every node attached to the line can see the conversations of the other nodes on the link.

Ring

In a ring topology LAN (shown in Figure 3.2), as in a bus topology LAN, all the nodes or devices in the network are attached to the network on the same cable or link. The difference is that a ring topology makes a complete circle. Both Token Ring/IEEE and Fiber Distributed Data Interface (FDDI) use a ring topology. FDDI is an American National Standard Institute (ANSI) X3T9.5 standard cable, which now supports up to Gigabit speeds using fiber- optic cabling. It can use a single ring for half-duplex operations or a dual “ring architecture for full-duplex operations.

Figure 3.2. A ring topology LAN. Notice that all of the nodes connect to the ring. Data for most implementations travels in one direction on the ring. However, many technologies, including Token Ring, allow for a second ring which allows for full-duplex operations.

When a break in the ring occurs, such as a cut cable or other cabling problem, it affects all the stations. This means that none of the stations connected can receive or transmit data. The longer the cable or link and the more attached stations, the more repeaters that are needed. However, due to timing distortions within signals, a limited number of repeaters can be used in the same network. In a ring-topology network, centralized access means that faults are easy to detect and isolate. Multiple rings are sometimes used to make a very robust and reliable network.

Star

The star topology is the most common topology in today's networks, and includes Ethernet, Fast Ethernet, and Gigabit Ethernet. Each node in a star topology connects to a dedicated link where the other end connects to a switch or hub. In the star-topology network shown in Figure 3.3, multiple devices are connected to a switch or hub.

Figure 3.3. A Star topology LAN. Notice that each workstation is directly connected to a hub or switch.

One of the best reasons to use a star topology is that a loss of any node will not disrupt network operations. It is also easy to add or remove a node from the network. From wiring to installation, it is particularly easy to set up a star topology network.