Review of LAN types and topologies

Home - Network - Lec1 and 2 - Lec3 - Lec3b - Lec4 - Lec5 and 6 - Lec7 - Lec8 - Lec10 - Lec11 - Lec12


A necessary prerequisite to this layered dissection of a LAN is an exploration of two other attributes: resource access methodology and topology. A LAN's resource access methodology describes the manner in which network-attached resources are shared. Frequently, this aspect of a network is referred to as its type. The two prevailing types are peer-to-peer and client/server.

A LAN's topology refers to its physical arrangement of hubs and wiring. The basic topologies include bus, star, ring, and switched.

Together, these attributes help form the context for your detailed exploration of a LAN's functional layers. This chapter explores all the possible permutations of both LAN types and topologies. Their benefits, limitations, and possible uses are also presented.

LAN-Attached Devices

Before we delve into LAN types and topologies, it's beneficial to first examine some of the basic resources that can be found on a LAN. The three most common primary devices are clients, servers, and printers. A primary device is one that can either directly access other devices or be accessed by other devices.

A server is any LAN-attached computer that hosts resources shared by other LAN-attached devices. A client is any computer that accesses resources stored on servers, via the LAN. Printers, of course, are output devices that produce hard copies of files. Numerous other devices, such as CD-ROM drives and tape archives, can also be accessed via a LAN, but they tend to be secondary resources. That is, they're connected to a primary device. Subordination of a device to another device, such as a CD-ROM drive to a server, is known as slaving. Printers, too, can be slaved off a primary device, or they can be primary devices directly connected to the network. Figure below illustrates the primary resources found in a LAN as well as the relationship of primary to secondary resources.



Types of Servers

Server is a word frequently used generically to describe all multiuser computers. It's important to note, however, that servers are a rather heterogeneous group. They're frequently specialized by function and described with an adjective. For example, there are file servers, print servers, and application servers.

File Servers
One of the most basic and familiar of the specialized servers is the file server. The file server is a centralized storage mechanism for files needed by a group of users. Placing these files in one centralized location, rather than scattered across numerous client-level machines, imparts several benefits. These benefits include the following:

Centralized location: All users enjoy a single, constant repository for shared files. This provides a dual benefit. Users don't have to search multiple potential storage locations to find a file; instead, files are stored in one place. Users also are relieved of the burden of maintaining separate logon credentials to multiple machines. One logon provides them with access to all the files they require.

Electric power conditioning: The use of a centralized server for file storage also enables the introduction of many techniques that can provide protection for data from inconsistencies in electric power. Fluctuations in the frequency of electricity, or even the sudden loss of power, can damage both a computer's data and hardware. Power filtration and battery backup through an uninterruptible power supply (UPS) is cost effective on a single server. Similar protection in a peer-to-peer network might be cost prohibitive because of the numbers of computers that would need protection.

Consistent data archiving: Storing all shared files in a common, centralized location greatly facilitates backups, because only a single output device and routine are required. Decentralized storage of data (at every desktop, for example) means that every desktop's data must be backed up separately. Backups are an essential protection against lost or damaged files. Suitable devices for backups include tape drives, writable optical drives, and even hard drives. Multiple hard drives can also be used in a technique known as striping. Striping involves multiple, simultaneous writes to different hard drives. Although this is primarily done to provide faster reading of data, it can also be used to create redundancy with every write operation,

Speed: The typical server is a more robust and fully configured platform than the typical client computer. This directly translates into a demonstrable performance gain relative to retrieving files in a peer-to-peer network.

The use of a file server does not always yield an increase in speed. You can usually access files stored locally more quickly than files stored on a remote computer and retrieved over a LAN. The speed increase discussed here is relative to the speed with which files can be retrieved from other client-level machines in a peer-to-peer network, not the speed with which files can be retrieved from a local hard drive.

Print Servers
Servers can also be used to share printers among the users of a LAN. Although the costs of printers, especially laser printers, have decreased considerably since their introduction, most organizations would be hard pressed to justify one for every desktop. Instead, servers are used to share one or more printers among the user population. This doesn't mean that a PC used as a print server cannot also be used as a desktop machine. Both NetWare and Windows NT allow workstations to share the use of a printer attached to it.

Alternatively, the print server software can run on the file server and share printers attached to it. The third possibility is a dedicated print server device, usually a very small "black box" with an Ethernet port and one or more parallel ports.

The alternative to using a print server is to directly attach the printer to the LAN. Many printers can be configured with a network interface card (NIC),which enables them to be directly attached to the LAN. This enables the printers to become print queue servers. Connecting printers directly to the LAN works well for all but the most print-intensive operations.

A print server's only function is to accept print requests from all networked devices, put them in a queue, and spool (send) them to the appropriate printer. This arrangement is illustrated in Figure below

Note: Although the term spool has become synonymous with printing, it's actually an acronym. SPOOL originally meant Simultaneous Peripheral Operations On Line. It's the temporary storage of programs and/or data in the form of output streams on magnetic media for later output or execution.

Each printer connected to a print server has its own queue (or waiting list). These queues represent the pecking order of all requests that are temporarily stored and waiting their turns to print. Requests are generally processed in the order they were received, although some network operating systems do allow for print jobs to be prioritized in other ways (so you could ensure that your boss's jobs went immediately to the top of the queue!).

Application Servers
Servers are also frequently used to run application software. Application servers, although superficially similar to file servers, are unique creatures. An application server hosts executable application software. To run that application software, a client must establish a connection across a network to the server. The application runs on that server. Servers that enable clients to download copies of an application for execution on their desktop computers are file servers. Their files are actually application software files, but they're functioning as file servers. The distinction isn't where the file is stored, but where the application executes. Application servers can enable an organization to reduce its overall cost of application software. Purchase prices and maintenance of a single, multiuser copy of an application are usually much less than the costs of acquiring and maintaining separate copies for each desktop.

Although it's usually desirable to separate application software from its data files by using separate servers (for example, an application server and a file server), there's one important exception to this rule. Because some applications build and maintain large relational databases, these applications and their databases should reside together on the application server.

The reason for this is simple: The mechanics of retrieving data from a database is very different from simply pulling down a Word or Excel file. The relational database application releases only the data requested and keeps everything else in its database. Office automation applications, such as Word and Excel, store information in standalone files that, typically, are not interdependent with other data in a complex structure. The relational database application is directly responsible for the integrity of the database and its indices. Managing the database across a network increases the risk of corrupting the indices and disabling the application.

Network Type
The network's type describes the manner in which attached resources can be accessed. Resources can be clients, servers, or any devices, files, and so on that reside on a client or server. These resources can be accessed in one of two ways: via peer-to-peer networks or server-based networks.

Peer-to-Peer Networks
A peer-to-peer network supports unstructured access to network-attached resources. Each device in a peer-to-peer network can be a client and a server simultaneously. All devices in the network are capable of accessing data, software, and other network resources directly. In other words, each networked computer is a peer of every other networked computer; there is no hierarchy.


Benefits
There are four main benefits to having a peer-to-peer network:

Limitations
Peer-to-peer networking is not without its risks and faults. Some of the more serious of these limitations are in the areas of security, performance, and administration. The peer-to-peer network suffers from numerous security weaknesses: Although the administrative burden is less in a peer-to-peer network than in a client/ server network, this burden is spread across users. This creates some logistical issues. Here are two of the gravest:

Uncoordinated, and probably highly inconsistent, backups of data and software. Each user is responsible for his own machine, so it's possible, and even likely, that each is going to perform backups at his own leisure.

Decentralized responsibility for enforcing filenaming conventions and storage locations. Given that there's no central repository for stored information or ally other logic by which LAN-attached resources are organized, keeping current with what information is stored where can be quite challenging. As with everything else in a peer-to-peer network, the effectiveness of the whole is directly dependent upon the degree to which methods and procedures are adhered to by all participants.

Lastly, performance also suffers. An integral aspect of a peer-to-peer network is that each machine is a multiuser machine. The typical machine is better suited for use as a singleuser client-only computer than it is for multiuser support. Consequently, the performance

of any given machine suffers noticeably, as perceived by its primary user, whenever remote users log on and share its resources.

The availability of files and any other resources that a given peer may host are only as available as that host. In other words, if a machine's primary user is out of the office and left it powered down, its resources are unavailable to the rest of the networked computers. This can be circumvented by leaving all machines powered on all the time, but doing so raises questions about other issues, such as security.

Another, more subtle aspect of performance is scalability. The peer-to-peer methodology is inherently nonscalable. The more peers are networked together, the more unmanageable the network becomes.

Uses
Peer-to-peer networking has two primary uses. First, it's ideally suited for small organizations with a limited budget for information technologies and limited need for information sharing. Alternatively, workgroups within larger organizations can also use this methodology for a tighter sharing of information within a particular group.

Server-Based Networks
Server-based networks introduce a hierarchy designed to improve the manageability of a network's various supported functions as the size of the network scales upward. Often, server-based networks are referred to as client/server networks.

In a server-based network, frequently shared resources are consolidated onto a separate tier of computers, known as servers. Servers, typically, do not have a primary user. Rather, they're multiuser machines that regulate the sharing of their resources across the base of clients. In this type of network, clients are relieved of the burden of functioning as servers to other clients.

Benefits There are many benefits inherent in the server-based approach to accessing network resources. These benefits directly correspond to the limitations of a peer-to-peer network. The areas of benefit are security, performance, and administration.

Server-based networks can be made, and kept, much more secure than peer-to-peer networks. Multiple factors contribute to this. First, security is managed centrally. Networked resources are no longer subjected to the "weakest link in the chain" theory that's an integral part of a peer-to-peer network.

Instead, all user accounts (also known as IDs) and passwords are centrally managed and verified before any user is granted access to requested resources. Coincidentally, this also makes the lives of the users better, by diminishing the need for multiple passwords.

Another benefit of this centralization of resources is that administrative tasks, such as backups, can be done consistently and reliably.

Server-based networks offer improved performance for networked computers in several ways. First, each client is relieved of the burden of processing requests from other clients for its stores. Each client in a server-based network need only keep up with the requests generated by its primary or only user.

More significantly, this processing is offloaded onto a server whose configuration is optimized for that service. Typically, a server contains more processing power, more memory, and larger, faster disk drives than those found in a client computer. The net effect is that users' client computers are able to better satisfy their own requests, and requests for resources centralized on a server are fulfilled much more effectively. Users, too, are spared the effort that would otherwise be required to learn which resources are stored where in a network. In a server-based network, the possible "hiding places" are reduced to just the number of servers on the network. In a server environment, server-based resources can be linked to as a logical drive. After the network drive linkage is established, remote resources stored on the server can be accessed as easily as any that are locally resident on a user's PC.

A server-based network is also very scalable. Regardless of how many clients are connected to the network, the resources are always centrally located. In addition, these resources are always centrally managed and secured. Consequently, the performance of the aggregate network isn't compromised by increases in scale.

Limitations
The server-based network has one limitation: It costs much more to implement and operate than a peer-to-peer network. There are many facets of this significant cost difference.

First, the hardware and software costs are significantly increased because of the need for a separate, networked computer that services the clients. Servers can be fairly sophisticated-which translates into expensive-machines.

The costs of operating a server-based network are also much higher. This is due to the need to have a trained professional administer the network and its servers. In a peer-topeer network, each user is responsible for the maintenance of his own machine; no individual needs to be dedicated to this function.

The last aspect is the potential cost of downtime. In a peer-to-peer network, the loss of any given peer translates into only a modest decrement in the available resources on the LAN. In a server-based LAN, the loss of a server can directly, and significantly, impact virtually all the users of the network. This increases the potential business risks of a server-based network. Numerous approaches, including clustering servers for redundancy, can be used to combat this risk. Unfortunately, every one of these approaches only serves to further drive up the cost of a server-based network.

Uses
Server-based networks are extremely useful in large organizations. They can also be useful in any circumstances that warrant tighter security or more consistent management of network-attached resources. The added cost of server-based networks, however, might place them beyond the reach of very small organizations.

Combination Networks
The distinctions between peer-to-peer and server-based networking aren't quite so clear as the preceding sections might suggest. They were presented as distinct types intentionally, as well as for academic purposes. In reality, the distinctions between them have been bluffed through the capabilities of numerous operating systems, such as Microsoft's Windows for Workgroups, Windows 95/98, and Windows NT/2000.

The norm today is a combination of peer-to-peer and server-based resource access in a single network. An example of this is a network with a server-based architecture that centralizes resources that are universally needed. Within this context, local workgroups can optionally provide peer-based access among themselves.

Note: Just as there's no single "best" network operating system, there's also no best type or best topology. Every enterprise, network, or administrator is different. Each has strengths and weaknesses, and each has unique needs. Thorough planning for your network and an understanding of your enterprise's needs are at least as important as, if not more than, being able to enumerate all the varieties of types and topologies. Remember, too, that no one has a better understanding of these needs than the people who use the network daily, so listen to your users at least as much as you listen to vendors.

LAN Topologies
Local area network topologies can be described using either a physical or a logical perspective. A physical topology describes the geometric arrangement of components that make up the LAN. The topology is not a map of the network. It's a theoretical construct that graphically conveys the shape and structure of the LAN.

A logical topology describes the possible connections between pairs of networked endpoints that can communicate. This is useful in describing which endpoints can communicate with which other endpoints and whether those pairs capable of communicating have a direct physical connection to each other. This chapter focuses only on physical topological descriptions.

Until recently, there have been three basic physical topologies: bus, ring, and star. Each basic topology is dictated by the physical LAN technology selected. For example, Token Ring networks, by definition, have historically used ring topologies. However, MSAUs (Token Ring's hubs, known more properly as Multistation Access Units), blurred the distinction between a ring and star topology for Token Ring networks. The result is known as a star ring. Similarly, the introduction of LAN switching is, again, changing the perception of topology. Switched LANs, regardless of frame type or access method, are topologically similar. The ring that used to exist at the electronics level within Token Ring's MSAUs no longer interconnects all the devices connected to that hub. Instead, each enjoys its own minimum ring that's populated with just two devices: the station device and the switch port. Consequently, switched should now be added to the long-standing triad of basic LAN topologies as a distinct fourth topology.

Tip: Switches implement a star topology, without regard for the Data Link Layer protocol for which they're designed. Given that the word switch has become readily understood (thanks to the tireless marketing campaigns of switch manu- facturers!), it has become more descriptive and readily understood than star bus or star ring. Consequently, switching can be regarded as a topology unto itself. For the purposes of any exam that might be lurking in your not-too-distant future, however, you might want to remember the terms star bus and star ring.

Switching has decoupled this historic coupling of topology and LAN technology: Literally all LAN technologies can be purchased in a switched implementation. This has significant ramifications for network access and, consequently, overall network performance. These ramifications are explored in more detail in the section titled "Switched Topology."

Note: Even though switches can be purchased to support any LAN type, including Ethernet, Token Ring, FDDI, and so forth, they're not translating bridges. That is, they're incapable of switching frames between dissimilar LAN architectures.

Bus Topology
A bus topology features all networked nodes interconnected, peer to peer, using a single, open-ended cable. This cable can support only a single channel. The cable is called the bus. Some bus-based technologies use more than a single cable. Consequently, they can support more than one channel, although each cable remains limited to just one transmission channel.

Both ends of the bus must be terminated with a resistive load, known as a terminating resistor. These resistors serve to prevent signal bounce. Whenever a station transmits, the signal that it puts on the wire automatically propagates in both directions. If a terminating resistor is not encountered, the signal reaches the end of the bus and reverses direction. Consequently, a single transmission can completely usurp all available bandwidth and prevent any other stations from transmitting. An example of bus topology is illustrated in Figure below

The typical bus topology features a single cable, supported by no external electronics, that interconnects all networked nodes in a peer-to-peer fashion. All connected devices listen to the bussed transmissions and accept those packets addressed to them. The lack of any external electronics, such as repeaters, makes bus LANs simple and inexpensive. The downside is that it also imposes severe limitations on distances, functionality, and scalability.

This topology is impractical for all but the smallest of LANs. Consequently, today's commercially available LAN products that use a bus topology are inexpensive peer-topeer networks that provide basic connectivity. These products are targeted at home and small office environments.

One exception to this was the IEEE's 802.4 Token Bus LAN specification. This technology is fairly robust, deterministic, and bears many similarities to a Token Ring LAN. Deterministic LANs offer the administrator a high degree of control in determining the maximum amount of time a frame of data can be in transmission. The primary difference, obviously, was that Token Bus was implemented on a bus topology.

Token Bus found extremely limited market support. Its implementation tended to be limited to factory production lines. Bus topologies, in general, prospered in myriad other forms. Two early forms of Ethernet, 10Base2 and 10Base5, used a bus topology and coaxial cabling. Buses also became a critical technology for interconnecting system-level components and peripheral devices within the internal architectures of computers.

Ring Topology
The ring topology started out as a simple peer-to-peer LAN topology. Each networked workstation had two connections: one to each of its nearest neighbors (see Figure 5.6). The interconnection had to form a physical loop, or ring. Data was transmitted unidirectionally around the ring. Each workstation acted as a repeater, accepting and responding to packets addressed to it and forwarding the other packets onto the next workstation on the ring.

The original LAN,ring topology featured peer-to-peer connections between workstations. These connections had to be closed; that is, they had to form a ring. The benefit of such LANs was that response time was fairly predictable. The more devices there were in the ring, the longer the network delays. The drawback was that early ring networks could be completely disabled if one of the workstations failed.

These primitive rings were made obsolete by IBM's Token Ring, which was later standardized by the IEEE's 802.5 specification. Token Ring departed from the peer-to-peer interconnection in favor of a repeating hub. This eliminated ring networks' vulnerability to workstation failure by eliminating the peer-to-peer ring construction. Token Ring networks, despite the name, are implemented with a star topology and a circular access method, as shown in Figure below


LANs can be implemented in a star topology, yet can retain a circular access method. The Token Ring network illustrated in Figure below demonstrates the virtual ring formed by the round-robin access method. The solid lines represent physical connections, and the dashed line represents the logical flow of regulated media access.


Functionally, the access token passes in a circular sequence, round-robin fashion, among the networked endpoints, even though they're all interconnected to a common hub. Therefore, many people succumb to the temptation of describing Token Ring networks as having a "logical" ring topology, even though they're shaped like a star. Evidence of this is found in Microsoft's Networking Essentials course and exam, which regard Token Ring as having a ring, not a star topology. In fact, the Token Ring hub, known properly as a Multistation Access Unit (MSAU), provides a physical ring internally, at the electronics level.

Star Topology

Star topology LANs have connections to networked devices that "radiate" out from a common point (that is, the hub, as shown in Figure above). Unlike ring topologiesphysical or virtual-each networked device in a star topology can access the media independently. These devices have to share the hub's available bandwidth. An example of a LAN with a star topology is 10BaseT Ethernet.

A small LAN with a star topology features connections that radiate out from a common point. Each connected device can initiate media access independent of the other connected devices.

Star topologies have become the dominant topology type in contemporary LANs. They're flexible, scalable, and relatively inexpensive compared to more sophisticated LANs with strictly regulated access methods. Stars have all but made buses and rings obsolete in LAN topologies and have formed the basis for the final LAN topology: switched.

Switched Topology
A switch is a multiport, Data Link Layer (OSI Reference Model Layer 2) device. A switch "learns" Media Access Control (MAC) addresses and stores them in an internal lookup table. Temporary, switched paths are created between the frame's originator and its intended recipient, and the frames are forwarded along that temporary path.

The typical LAN with a switched topology is illustrated in Figure below. It features multiple connections to a switching hub. Each port, and the device to which it connects, has its own dedicated bandwidth. Although originally switches forwarded frames based on the frames' MAC addresses, technological advances are rapidly changing all that. Switches are available (called "Layer 3" switches) that can process cells, frames, and even packets that use a Layer 3 address like IP.


Note: A frame is a variable-length structure that contains data, source, and destination addresses as well as other data fields required for its carriage and forwarding in Layer 2 of the OSI Reference Model. Cells are very similar to frames, except they feature a fixed, not variable, length. Packets are a construct of protocols that operate at Layer 3 of the OSI Reference Model.

Switches can improve the performance of a LAN in two important ways. First, they increase the aggregate bandwidth available throughout the network. For example, a switched Ethernet hub with eight ports contains eight separate collision domains of 10Mbps each, for an aggregate of 80Mbps of bandwidth.

The second way that switches improve LAN performance is by reducing the number of devices forced to share each segment of bandwidth. Each switch-delineated collision domain is inhabited by only two devices: the networked device and the port on the switching hub to which it connects. These are the only two devices that can compete for the 10Mbps of bandwidth on the segment. In networks that do not utilize a media access method that's based on competition for bandwidth-such as Token Ring and FDDI-the tokens circulate among a much smaller number of networked machines than are typically supported in competition-based networks.

One area for concern with large switched implementations is that switches do not isolate broadcasts. They bolster performance solely by segmenting collision, not broadcast, domains. Excessive broadcast traffic can significantly and adversely impact LAN performance.

Complex Topologies
Complex topologies are extensions and/or combinations of basic physical topologies. Basic topologies, by themselves, are adequate for only very small LANs. The scalability of the basic topologies is extremely limited. Complex topologies are formed from these building blocks to achieve a custom-fitted, scalable topology.

Daisy Chains
The simplest of the complex topologies is developed by serially interconnecting all the hubs of a network, as shown in Figure below. This is known as daisy-chaining. This simple approach uses ports on existing hubs for interconnecting the hubs. Therefore, no incremental cost is incurred during the development of such a backbone.

Small LANs can be scaled upward by daisy-chaining hubs together. Daisy chains are easily built and don't require any special administrative skills. Daisy chains were, historically, the interconnection method of choice for emerging, first-generation LANs.

The limits of daisy-chaining can be discovered in a number of ways. LAN technology specifications, such as 802.3 Ethernet, dictate the maximum size of the LAN in terms of the number of hubs and/or repeaters that may be strung together in sequence. The distance limitations imposed by the Physical Layer, multiplied by the number of devices, dictate the maximum size of a LAN. This size is referred to as a maximum network diameter. Scaling beyond this diameter adversely affects the normal functioning of that LAN. Maximum network diameters frequently limit the number of hubs that can be interconnected in this fashion. This is particularly true of contemporary high-performance LANs, such as Fast Ethernet, that place strict limitations on the network diameter and the number of repeaters that can be strung together.

Note: A repeater is a device that accepts an incoming signal, amplifies it back to its original volume, and places it back on the network. Typically, signal amplification and repetition functions are incorporated into hubs. Consequently, the two terms (repeaters and hubs) can be used synonymously

Daisy-chaining networks that use a contention-based media access method can become problematic long before network diameter is compromised, however. Daisy-chaining increases the number of connections-and therefore the number of devices--on a LAN. It does not increase aggregate bandwidth or segment collision domains. Daisy-chaining simply increases the number of machines sharing the network's available bandwidth. Too many devices competing for the same amount of bandwidth can create collisions and quickly incapacitate a LAN.

This topology is best left to LANs with less than a handful of hubs and little, if any, wide area networking.

Hierarchies
Hierarchical topologies consist of more than one layer of hubs. Each layer serves a different network function. The bottom tier is reserved for user station and server connectivity. Higher-level tiers provide aggregation of the user-level tier. In much simpler terms, many user-level hubs are interconnected via a lesser number of higher-level hubs. The hubs themselves can be identical devices; their only distinction lies in their application. A hierarchical arrangement is best suited for medium- to large-sized LANs that must be concerned with scalability of the network and with traffic aggregation.

Hierarchical Rings
Ring networks can be scaled up by interconnecting multiple rings in a hierarchical fashion, as shown in Figure below. User station and server connectivity can be provided by as many limited size rings as are necessary to provide the required level of performance. A second-tier ring, either Token Ring or FDDI, can be used to interconnect all the userlevel rings and to provide aggregated access to the wide area network (WAN).

Small ring LANs can be scaled by interconnecting multiple rings hierarchically. In Figure below, two distinct 16Mbps Token Rings, shown logically as loops, are used to interconnect the user stations, and separate FDDI loops are used for the servers and backbone tier.


Hierarchical Stars

Star topologies, too, can be implemented in hierarchical arrangements of multiple stars, as shown in Figure above Hierarchical stars can be implemented as a single collision domain or segmented into multiple collision domains using switches, routers, or bridges.

Note: A collision domain consists of all the devices that compete for the right to transmit on a shared media. Switches, bridges, and routers all segment the collision domain (that is, they create multiple, smaller collision domains).

A hierarchical star topology uses one tier for user and server connectivity and the second tier as a backbone.


Hierarchical Combinations
Overall network performance can be enhanced by not force-fitting all the functional requirements of the LAN into a single solution. Today's high-end switching hubs enable you to mix multiple technologies. New topologies can be introduced by simply inserting the appropriate circuit board into the multislot chassis of the switching hub. A hierarchical topology lends itself to combinations of topologies, as shown in Figure below.

In this example of a hierarchical combination topology, an Asynchronous Transfer Mode (ATM) backbone is used to interconnect the user-level hubs. FDDI interconnects the server farm (that is, a group of servers isolated on their own segment); Ethernet interconnects the user stations. This approach differentiates the LAN into functional components (station-connect, server-connect, and backbone) and enables the ideal technology for each function to be used. These functional areas are addressed further in the next section, "LAN Functional Areas."

LAN Functional Areas
Topological variation can be an important way to optimize network performance for each of the various functional areas of a LAN. LANs contain four distinct functional areas: station connectivity, server connectivity, WAN connectivity, and backbone. Each may be best served by a different basic or complex topology.

Station Connectivity
The primary function of most LANs is station connectivity. Station connectivity, as its name implies, is the portion of the LAN used to connect the user stations to the network. This functional area tends to have the least stringent performance requirements of the LAN's functional areas. There are obvious exceptions to this, such as CAD/CAM workstations, desktop videoconferencing, and so on. In general, though, compromises in the cost and performance of this part of a LAN's technology and topology are less likely to adversely affect the network's performance.

Providing connectivity to machines that have divergent network performance requirements may require the use of multiple LAN technologies, as shown in Figure below. Fortunately, many of today's hub manufacturers can support multiple technologies from the same hub chassis.


LANs provide basic connectivity to user stations and the peripherals that inhabit them. Differences in the network performance requirements of user station equipment can necessitate a mixed topology/technology solution.

Server Connectivity
Servers tend to be much more robust than user workstations. Servers tend to be a point of traffic aggregation and must serve many clients and/or other servers. In the case of highvolume servers, this aggregation must be designed into a LAN's topology; otherwise, clients and servers suffer degraded network performance. Network connectivity to servers, typically, should also be more robust than station connectivity in terms of the available bandwidth and the robustness of the access method.

LAN topologies can also be manipulated to accommodate the robust network performance requirements of servers and server clusters. In Figure above for example, a hierarchical combination topology is employed. The server farm is interconnected with a small FDDI loop; the less robust user stations are interconnected with Ethernet.

WAN Connectivity
A frequently overlooked aspect of a LAN's topology is its connection to the wide area network (WAN). In many cases, WAN connectivity is provided by a single connection from the backbone to a router, as shown in Figure below


The LAN's connection to the router that provides WAN connectivity is a crucial link in a building's overall LAN topology. Improper technology selection at this critical point can result in unacceptably deteriorated levels of performance for all traffic entering or exiting the building's LAN. LAN technologies that use a contention-based access method are highly inappropriate for this function.

Networks that support a large quantity of WAN-to-LAN and LAN-to-WAN traffic benefit greatly from having the most robust connection possible in this aspect of their overall topology. The technology selected should be robust in terms of its nominal transmission rate and its access method. Contention-based technologies should be avoided at all costs. The use of a contention-based media, even on a dedicated switched port, may become problematic in high-usage networks. This is the bottleneck for all traffic coming into and trying to get out of the building's LAN.

Backbone Connectivity
A LAN's backbone is the portion of its facilities used to interconnect all the hubs. A backbone can be implemented in several topologies and with several different network components, as shown in Figure below

The LAN's backbone provides a critical function: It interconnects all the locally networked resources and, if applicable, the WAN. The logical depiction of a backbone, as shown in Figure above, can be implemented in a wide variety of ways.

Deter-mining which backbone topology is correct for your LAN is not easy. Some options are easier to implement, very affordable, and easy to manage. Others can be more costly to acquire and operate. Another important difference lies in the scalability of the various backbone topologies. Some are easy to scale, up to a point, and then require reinvestment to maintain acceptable levels of performance. Each option must be examined individually relative to your particular situation and requirements.

Serial Backbone
A serial backbone, shown in Figurebelow, is nothing more than a series of hubs daisychained together. As described in the preceding section, this topology is inappropriate for all but the smallest of networks.


Distributed Backbone
A distributed backbone is a form of hierarchical topology that can be built by installing a backbone hub in a central location. A building's PBX room (the telephone closet) usually serves as the center of its wiring topology. Consequently, it's the ideal location for a distributed backbone hub. Connections from this hub are distributed to other hubs throughout the building, as shown in Figure below


A distributed backbone can be developed by centrally locating the backbone hub. Connections are distributed from this hub to other hubs throughout the building. Unlike the serial backbone, this topology enables LANs to span large buildings without compromising maximum network diameters.

If you consider a distributed backbone, make sure you understand the building's wire topology and the distance limitations of the various LAN media choices. In medium-to-large locations, the only viable option for implementing a distributed backbone is probably fiber-optic cabling.

Collapsed Backbone
A collapsed backbone topology features a centralized router that interconnects all the LAN segments in a given building. The router effectively creates multiple collision and broadcast domains, thereby increasing the performance of each of the LAN segments.

Routers operate at Layer 3 of the OSI Reference Model. They are incapable of operating as quickly as hubs. Consequently, they can limit effective throughputs for any LAN traffic that originates on one LAN segment but terminates on another.

Collapsed backbones, like the one shown in Figure below, also introduce a single point of failure in the LAN. This is not a fatal flaw. In fact, many of the other topologies also introduce a single point of failure in the LAN. Nevertheless, that weakness must be considered when planning a network topology.


An important consideration in collapsed backbone topologies is that user communities are seldom conveniently distributed throughout a building. Instead, users are scattered far and wide. This means that there's a good chance they will be found on both sides of the LAN's collapsed backbone router. Subsequently, simple network tasks among the members of a workgroup are likely to traverse the router. Care should be taken when designing collapsed backbone LANs to absolutely minimize the amount of traffic that must cross the router. Use it as a traffic aggregator for LAN-level resources, such as WAN facilities, and not indiscriminately as a bridge.

Parallel Backbone
In some of the cases where collapsed backbones are an untenable solution, a modified version may prove ideal. This modification is known as the parallel backbone. The reasons for installing a parallel backbone are many. Here are some examples: Regardless of the reason, running parallel connections from a building's collapsed backbone router to the same telephone closet enables supporting multiple segments to be run from each closet, as shown in Figure below


The Parallel backbone topology is a modification of the collapsed backbone. Multiple segments can be supported in the same telephone closet or equipment room. This marginally increases the cost of the network, but can increase the performance of each segment and satisfy additional network criteria, such as security.

A careful understanding of the performance requirements imposed by customers, stratified by LAN functional areas, is the key to developing the ideal topology for any set of user requirements. The potential combinations are limited only by your imagination. Continued technological innovation will only serve to increase the topological variety available to network designers.

Note: Many of the complex topologies presented in this chapter are for your edification only. They are of more practical use on the job than they are during certification exams. In fact, vendors frequently refer to many of these complex topologies by different names. Rather than get caught up in names, try to graps their concepts strengths and weaknesses.

Summary
An important aspect of a LAN is the manner in which it supports resource access. Although this is more a function of the network operating systems than it is LAN hardware, it directly impacts the traffic flow and performance of the LAN.

LAN topology, too, is directly related to the efficacy of your LAN. The four basic topologies (bus, star, ring, and switched) can be mixed and matched, stacked or linked, in an almost infinite variety. Understanding the benefits and limitations of the basic topologies, as well as using them wisely, is critical to the selection of one that best satisfies the expected demands.

The topics presented in this chapter should reinforce the fact that there's more to networks than just their hardware and wiring. The way the physical components are arranged and the manner in which attached resources are accessed are equally important in technology selection.