Topology is the physical and logical arrangement of a network. The physical arrangement of the network refers to how the workstations, servers, and other equipment are joined together with cables and connectors. The logical arrangement of a network refers to how the workstations, servers, and other equipment relate to each other in terms of traffic flow. There are three primary LAN topologies: linear bus, ring, and star. Another network topology is hierarchical in nature, which may incorporate elements of the bus, ring, and star. The appropriate physical and logical topology for a LAN is determined by reliability and cost objectives as well as by the connectivity requirements of users.
In a linear bus topology, stations are arranged along a single length of cable, which can be extended at either end or at both ends to accommodate more nodes (Figure 65). The network consists of coaxial cable, such as the RG-58 A/U cable used with 10Base2 Ethernet LANs. The nodes are attached to the cable with a BNC (Bayonet Nut Connector) T-connector (Figure 66), the stem of which attaches to the network interface card (NIC). A BNC barrel connector attaches cable segments and a BNC terminator connector caps the cable ends. Of course, twisted pair wiring is most often used for Ethernet LANs, in which case RJ45 connectors provide the connections between devices.
The linear bus topology.
A linear bus network can be further extended. For example, a tree topology is actually a complex linear bus in which the cable branches at either or both ends, but offers only one transmission path between any two stations.
In a ring topology, nodes are arranged along the transmission path so data passes through each successive station before returning to its point of origin. As its name implies, the ring topology consists of nodes that form a closed circle (Figure 67).
In token-ring LANs, a small packet called a token is circulated around the ring, giving each station in sequence a chance to put information on the network. The station seizes the token, replacing it with an information frame. Only the addressee can claim the message. At the completion of pass through the central node, which acts as a processing and coordinating point for the network. This central node is generally referred to as a hub. Information addressed to one or more specific nodes is sent through the central node and switched to the proper receiving station(s) over a dedicated physical path.
BNC T-connectors are used to connect two cable segments to a node’s network interface card (NIC).
The ring topology.
More complex LAN topologies can be created from the basic bus, ring, and star topologies. One of these is the “dual ring of trees” on Fiber Distributed Data Interface (FDDI) networks that is created with special categories of equipment. These equipment types may be arranged in any of three topologies: dual ring, tree, and dual ring of trees (Figure 69).
With FDDI, a dual ring of trees can be used to create a hierarchical topology to enhance network reliability.
In the dual ring topology, dual attached stations (DASs) form a physical loop, where all the stations are dual attached. In a tree topology, remote single attached stations (SASs) are linked to a concentrator, which is connected to another concentrator on the main ring.
Any DAS connected to a concentrator performs as a SAS. Concentrators can be used to create a network hierarchy, which is known as a dual ring of trees. This topology offers a flexible hierarchical system design that is efficient and economical. Devices requiring highly reliable communications attach directly to the main ring, while those that are less crucial attach to branches off the main ring. Thus, SAS devices can communicate with the main ring, but without the added cost of equipping them with a dual-ring interface or a loop-around capability that would otherwise be required to ensure the reliability of the ring in the event of a station failure.
Each topology has advantages and disadvantages. The bus topology characteristic of Ethernet LANs is the most economical and easiest to install. The ring is slightly more expensive and complicated. In both types of topologies, when one node malfunctions or becomes inoperable, the nodes on either side of it cannot communicate. This can be overcome by adding a hub. The nodes communicate with each other over separate cable segments via the collapsed backbone within the hub. If one node become inoperable, the other nodes are not affected since they are no longer directly connected.
In the case of Ethernet, although the physical topology has changed from a linear bus to a star, the logical operation remains unchanged in that Ethernet’s Carrier Sense Multiple Access with Collision Detection (CSMA/CD) protocol still governs access. In the case of token ring, although the physical topology has changed from a ring to a star, the logical operation remains unchanged in that token ring’s circulating “token” still governs access.
When it comes to link availability, the star topology is highly reliable. In this topology, all network devices connect to a central hub through dedicated or shared LAN segments. Although the loss of a link prevents communication between the hub and the affected node(s), all other nodes continue to operate as before unless the hub itself suffers a catastrophic failure.
To ensure a high degree of reliability, the hub has redundant control logic, backplane, and power supply. The hub’s management system can enhance the fault tolerance of these redundant subsystems by monitoring their operation and reporting any problems. With the power supply, for example, monitoring may include hotspot detection and fan operation to detect trouble before it disrupts hub operation. Upon the failure of the main power supply, the redundant unit switches over automatically or manually under the network manager’s control without disrupting the network. If a fan goes out, an alarm can be sent to the management console as well as to a technician’s pager.
The flexibility of the hub architecture lends itself to varying degrees of fault tolerance, depending on the importance of the applications. For example, workstations running financial modeling applications may share a link to the same LAN module at the hub. Although this configuration might seem economical, it is problematic in that a failure in the LAN module will put all of the workstations on that link out of commission.
A slightly higher degree of fault tolerance can be achieved by distributing the workstations among two LAN modules and links. That way, the failure of one module will affect only half of the workstations. A one-to-one correspondence of workstations to modules offers an even greater level of fault tolerance in that the failure of one module impacts only the workstation connected to it. However, this configuration is also the most expensive solution.
A mission-critical application may demand the highest level of fault tolerance. This can be achieved by connecting the workstation to two LAN modules at the hub with separate links. The ultimate in fault tolerance can be achieved by connecting one of those links to a different hub. In this arrangement, a transceiver is used to split the links from the application’s host computer, enabling each link to connect with a different module in the hub or to a different hub. In each case, the physical topology changes, but the logical topology remains the same.
With the introduction of switching equipment into LANs, it is now possible to fine-tune the topology of smaller subsections of an organization’s network. Network planners can provide the advantages of one topology over another to meet the specific needs of individuals, workgroups, or departments.