Upgrading and Repairing PCs

Hardware Elements of Your Network

The choice of a data-link protocol affects the network hardware you choose. Because Ethernet, Fast Ethernet, Token-Ring, and other data-link protocols use different hardware, you must select the protocol before you can select appropriate hardware, including network interface cards, cables, and hubs or switches.

Network Interface Cards

On most computers, the network interface adapter takes the form of a network interface card (NIC) that fits into a PCI slot on a desktop computer or a PC Card (PCMCIA) slot on a notebook computer. Although network cards for older systems might use the ISA or EISA slot standards, these don't support high-speed network standards and are obsolete. Many recent systems incorporate the network interface adapter onto the motherboard, but this practice is more commonly found in workstations and portable computers and rarely in servers because most network administrators prefer to select their own NICs.

Note You can also adapt the USB port for Ethernet use with a USB-to-Ethernet adapter, but because USB 1.1 ports run at only 12Mbps (compared to Fast Ethernet running at 100Mbps), a performance bottleneck results. USB 2.0-to-Fast Ethernet adapters are now available from several vendors and are a suitable choice if your system has USB 2.0 ports but lacks a free PCI or PC Card slot.

Ethernet and Token-Ring adapters have unique hardware addresses coded into their firmware. The data link layer protocol uses these addresses to identify the other systems on the network. A packet gets to the correct destination because its data link layer protocol header contains the hardware addresses of both the sending and receiving systems.

10/100 Ethernet network adapters range in price from under $20 for client adapters to as much as $300 or more for server-optimized adapters. Token-Ring adapters are much more expensive, ranging in price from around $170 for client adapters to more than $500 for server-optimized adapters. Network adapters are built in all the popular interface-card types and are also optimized for either workstation or server use. For first-time network users, so-called network-in-a-box kits are available that contain two 10/100 Fast Ethernet NICs, a small hub or switch, and prebuilt UTP cables for less than $90. When combined with the built-in networking software in Windows, these kits make networking very inexpensive. A number of 10/100/1000-BASE-TX Gigabit Ethernet adapters for use with UTP cable can be purchased for under $60, and some 10/100 switches now also feature Gigabit Ethernet ports.

For client workstations (including peer servers on peer-to-peer networks), the following sections contain my recommendations on the features you need.

Speed

Your NIC should run at the maximum speed you want your network to support. For a Fast Ethernet network, for example, you should purchase Ethernet cards that support 100BASE-TX's 100Mbps speed. Most Fast Ethernet cards also support standard Ethernet's 10Mbps speed, allowing the same card to be used on both older and newer portions of the network. To verify dual-speed operation, look for network cards identified as 10/100 Ethernet.

Your NIC should support both half-duplex and full-duplex operation:

• Half-duplex means that the network card can only send or only receive data in a single operation.

• Full-duplex means that the network card can both receive and send simultaneously. Full-duplex options boost network speed if switches are used in place of hubs. For example, 100Mbps Fast Ethernet cards running in full-duplex mode have a maximum true throughput of 200Mbps, with half going in each direction.

Unlike hubs, which broadcast data packets to all computers connected to it, switches create a direct connection between the sending and receiving computers. Thus, switches provide faster performance than hubs; most switches also support full-duplex operation, doubling the rated speed of the network when full-duplex network cards are used.

Although, at one time, switches were hard to justify on a workgroup peer-to-peer LAN because of their extra cost, switches are now available for little more than the price of hubs with a similar number of ports and are recommended for LANs of any size. If you plan to use your Ethernet-based home or office network to share an Internet connection, consider purchasing a router with an integrated switch. This combination makes networking simple and requires less space than separate routers and switches. Many router/switch combinations cost little more than a switch or router would separately. You can also purchase a wireless access point and router with an integrated switch so both wired and wireless workstations can share a single Internet connection. For other types of networks, you can use a gateway to connect your network with the Internet.

Bus Type

If you are networking desktop computers built from 1995 to the present, you should consider only PCI-based NICs (these computers typically have three or more PCI slots). Although many older computers still have at least one ISA or combo ISA/PCI expansion slot, the superior data bus width and data transfer rate of PCI make it the only logical choice for networks of all types. The integrated NIC found on some recent PC motherboards is also a PCI device. Alternative interfaces include USB or PC Card/Cardbus adapters that are often used for portable systems.

Table 20.3 summarizes the differences between all the types of interfaces used by network cards.

Table 20.3. Bus Choices for Client PC NICs

Bus Type	Bus Width (Bits)	Bus Speed (MHz)	Data Cycles per Clock	Bandwidth (MBps)
8-bit ISA (AT)	8	8.33	1/2	4.17
16-bit ISA (AT-Bus)	16	8.33	1/2	8.33
EISA Bus	32	8.33	1	33
MCA-16 Streaming	16	10	1	20
MCA-32 Streaming	32	10	1	40
MCA-64 Streaming	64	10	1	80
PC-Card (PCMCIA)	16	10	1	20
CardBus	32	33	1	133
PCI	32	33	1	133
PCI 66MHz	32	66	1	266
PCI 64-bit	64	33	1	266
PCI 66MHz/64-bit	64	66	1	533
USB 1.1	1	12	1	1.5
USB 2.0	1	480	1	60
Note: ISA, EISA, and MCA are no longer used in current motherboard designs. MBps = Megabytes per second ISA = Industry Standard Architecture, also known as the PC/XT (8-bit) or AT-Bus (16-bit) EISA = Extended Industry Standard Architecture (32-bit ISA) MCA = Microchannel Architecture (IBM PS/2 systems) PC-Card = 16-bit PCMCIA (Personal Computer Memory Card International Association) interface CardBus = 32-bit PC-Card PCI = Peripheral Component Interconnect USB = universal serial bus

Although a few ISA-based NICs are still on the market, their slow speeds and narrow bus widths severely restrict their performance. Most ISA-based Ethernet cards can't support speeds above 10Mbps and thus don't support Fast Ethernet or Gigabit Ethernet. A few vendors make 10/100 Ethernet cards for the ISA slot, but their performance is substantially lower than PCI cards. If you are shopping for a network card for a laptop or notebook system, look for Cardbus types, which are significantly faster than PC Cards and those using USB.

Network Adapter Connectors

Ethernet adapters typically have a connector that looks like a large telephone jack called an RJ-45 (for 10BASE-T and Fast Ethernet twisted-pair cables), a single BNC connector (for Thinnet coaxial cables), or a D-shaped 15-pin connector called a DB15 (for Thicknet coaxial cables). A few older 10Mbps adapters have a combination of two or all three of these connector types; adapters with two or more connectors are referred to as combo adapters. Token-Ring adapters can have a 9-pin connector called a DB9 (for Type 1 STP cable) or sometimes an RJ-45 jack (for Type 3 UTP cable). Figure 20.3 shows all three of the Ethernet connectors.

The following figures provide profile views of the most common types of NIC connections. Figure 20.4 shows a 10BASE-2 NIC configured to be at the end of a network; the T-adapter connected to the BNC connector has a Thinnet (RG-58) cable attached to one side and a 50-ohm terminator at the other end.

Figure 20.5 shows a 10BASE-T NIC with its UTP cable attached.

Virtually all standard and 10/100 Ethernet NICs made for client-PC use on the market today are designed to support unshielded twisted-pair (UTP) cable exclusively; Gigabit Ethernet cards made for wired (not fiber-optic) networks also use only UTP cable. If you are adding a client PC to an existing network that uses some form of coaxial cable, you have three options:

• Purchase a combo NIC that supports coaxial cable as well as RJ-45 twisted-pair cabling.

• Purchase a media converter that can be attached to the coaxial cable to allow the newer UTP-based NICs to connect to the existing network.

• Use a switch or hub that has both coaxial cable and RJ-45 ports. A dual-speed (10/100) device is needed if you are adding one or more Fast Ethernet clients.

For maximum economy, NICs and network cables must match, although media converters can be used to interconnect networks based on the same standard, but using different cable.

Network Cables

Originally, all networks used some type of cable to connect the computers on the network to each other. Although various types of wireless networks are now on the market, most office and home networks are still based on one of the following wired topologies:

• Coaxial cable

• Twisted-pair cabling

Thick and Thin Ethernet Coaxial Cable

The first versions of Ethernet were based on coaxial cable. The original form of Ethernet, 10BASE-5, used a thick coaxial cable (called Thicknet) that was not directly attached to the NIC. A device called an attachment unit interface (AUI) ran from a DB15 connector on the rear of the NIC to the cable. The cable had a hole drilled into it to allow the "vampire tap" to be connected to the cable. NICs designed for use with thick Ethernet cable are almost impossible to find as new hardware today.

10BASE-2 Ethernet cards use a BNC (Bayonet-Neill-Concilman) connector on the rear of the NIC. Although the thin coaxial cable (called Thinnet or RG-58) used with 10BASE-2 Ethernet has a bayonet connector that can physically attach to the BNC connector on the card, this configuration is incorrect and won't work. Instead, a BNC T-connector attaches to the rear of the card, allowing a thin Ethernet cable to be connected to either both ends of the T (for a computer in the middle of the network) or to one end only (for a computer at the end of the network). A 50-ohm terminator is connected to the other arm of the T to indicate the end of the network and prevent erroneous signals from being sent to other clients on the network. Some early Ethernet cards were designed to handle thick (AUI/DB15), thin (RG-58), and UTP (RJ-45) cables. Combo cards with both BNC and RJ-45 connectors are still available but can run at only standard Ethernet speeds.

Figure 20.6 compares Ethernet DB-15 to AUI, BNC coaxial T-connector, and RJ-45 UTP connectors to each other, and Figure 20.7 illustrates the design of coaxial cable.

Twisted-Pair Cable

Twisted-pair cable is just what its name implies: insulated wires within a protective casing with a specified number of twists per foot. Twisting the wires reduces the effect of electromagnetic interference (that can be generated by nearby cables, electric motors, and fluorescent lighting) on the signals being transmitted. Shielded twisted pair (STP) refers to the amount of insulation around the cluster of wires and therefore its immunity to noise. You are probably familiar with unshielded twisted-pair (UTP) cable; it is often used for telephone wiring. Figure 20.8 shows unshielded twisted-pair cable; Figure 20.9 illustrates shielded twisted-pair cable.

Shielded Versus Unshielded Twisted Pair When cabling was being developed for use with computers, it was first thought that shielding the cable from external interference was the best way to reduce interference and provide for greater transmission speeds. However, it was discovered that twisting the pairs of wires is a more effective way to prevent interference from disrupting transmissions. As a result, earlier cabling scenarios relied on shielded cables rather than the unshielded cables more commonly in use today. Shielded cables also have some special grounding concerns because one, and only one, end of a shielded cable should be connected to an earth ground; issues arose when people inadvertently caused grounding loops to occur by connecting both ends or caused the shield to act as an antenna because it wasn't grounded. Grounding loops are situations in which two different grounds are tied together. This is a bad situation because each ground can have a slightly different potential, resulting in a circuit that has very low voltage but infinite amperage. This causes undue stress on electrical components and can be a fire hazard.

Most Ethernet and Fast Ethernet installations that use twisted-pair cabling use UTP because the physical flexibility and small size of the cable and connectors makes routing it very easy. However, its lack of electrical insulation can make interference from fluorescent lighting, elevators, and alarm systems (among other devices) a major problem. If you use UTP in installations where interference can be a problem, you need to route the cable away from the interference, use an external shield, or substitute STP for UTP near interference sources.

Network Topologies

Each computer on the network is connected to the other computers with cable (or some other medium). The physical arrangement of the cables connecting computers on a network is called the network topology.

The three types of basic topologies used in computer networks are as follows:

• Bus. Connects each computer on a network directly to the next computer in a linear fashion. The network connection starts at the server and ends at the last computer in the network.

• Star. Connects each computer on the network to a central access point.

• Ring. Connects each computer to the others in a loop or ring.

These different topologies are often mixed, forming what is called a hybrid network. For example, you can link the hubs of several star networks together with a bus, forming a star-bus network. Rings can be connected in the same way.

Table 20.4 summarizes the relationships between network types and topologies.

Table 20.4. Network Cable Types and Topologies

Network Type	Standard	Cable Type	Topology
Ethernet	10BASE-2	Thick coaxial	Bus
	10BASE-5	Thin (RG-58) coaxial	Bus
	10BASE-T	Cat 3 or Cat 5 UTP	Star
Fast Ethernet	100BASE-TX	Cat 5 UTP	Star
Gigabit Ethernet	1000BASE-TX	Cat 5 UTP	Star
Token-Ring	(all)	STP or coaxial	Logical Ring

The bus, star, and ring topologies are discussed in the following sections.

Bus Topology

The earliest type of network topology was the bus topology, which uses a single cable to connect all the computers in the network to each other, as shown in Figure 20.10. This network topology was adopted initially because running a single cable past all the computers in the network is easier and less wiring is used than with other topologies. Because early bus topology networks used bulky coaxial cables, these factors were important advantages. Both 10BASE-5 (thick) and 10BASE-2 (thin) Ethernet networks are based on the bus topology.

However, the advent of cheaper and more compact unshielded twisted-pair cabling, which also supports faster networks, has made the disadvantages of a bus topology apparent. If one computer or cable connection malfunctions, it can cause all the stations beyond it on the bus to lose their network connections. Thick Ethernet (10BASE-5) networks often fail because the vampire tap connecting the AUI device to the coaxial cable comes loose. In addition, the T-adapters and terminating resistors on a 10BASE-2 Thin Ethernet network can also come loose or be removed by the user, causing all or part of the network to fail. Another drawback of Thin Ethernet (10BASE-2) networks is that adding a new computer to the network between existing computers might require replacement of the existing network cable between the computers with shorter segments to connect to the new computer's network card and T-adapter.

Ring Topology

Another topology often listed in discussions of this type is a ring, in which each workstation is connected to the next and the last workstation is connected to the first again (essentially a bus topology with the two ends connected). Two major network types use the ring topology:

• Fiber Distributed Data Interface (FDDI). A network topology used for large, high-speed networks using fiber-optic cables in a physical ring topology

• Token-Ring. Uses a logical ring topology

A Token-Ring network resembles a 10BASE-T or 10/100 Ethernet network at first glance because both networks use a central connecting device and a physical star topology. Where is the "Ring" in Token-Ring?

The ring exists only within the device that connects the computers, which is called a multistation access unit (MSAU) on a Token-Ring network (see Figure 20.11).

Signals generated from one computer travel to the MSAU, are sent out to the next computer, and then go back to the MSAU again. The data is then passed to each system in turn until it arrives back at the computer that originated it, where it is removed from the network. Therefore, although the physical wiring topology is a star, the data path is theoretically a ring. This is called a logical ring.

A logical ring that Token-Ring networks use is preferable to a physical ring network topology because it affords a greater degree of fault tolerance. As on a bus network, a cable break anywhere in a physical ring network topology, such as FDDI, affects the entire network. FDDI networks use two physical rings to provide a backup in case one ring fails. By contrast, on a Token-Ring network, the MSAU can effectively remove a malfunctioning computer from the logical ring, enabling the rest of the network to function normally.

Star Topology

The most popular type of topology in use today has separate cables to connect each computer to a central wiring nexus, often called a hub or concentrator; a switch can also be used in place of a hub. Figure 20.12 shows this arrangement, which is called a star topology.

Because each computer uses a separate cable, the failure of a network connection affects only the single machine involved. The other computers can continue to function normally. Bus cabling schemes use less cable than the star but are harder to diagnose or bypass when problems occur. At this time, Fast Ethernet in a star topology is the most commonly implemented type of LAN; this is the type of network you build with most preconfigured network kits. 10BASE-T Ethernet and 1000BASE-TX Gigabit Ethernet also use the star topology. 10BASE-T can use either Category 3 or Category 5 UTP, whereas Fast Ethernet and Gigabit Ethernet require Category 5 UTP or greater.

Hubs and Switches for Ethernet Networks

As you have seen, modern Ethernet workgroup networks are based on UTP cable with workstations arranged in a star topology. The center of the star uses a multiport connecting device that can be either a hub or a switch. Although hubs and switches can be used to connect the network—and can have several features in common—the differences between them are also significant and are covered in the following sections.

All Ethernet hubs and switches have the following features:

• Multiple RJ-45 UTP connectors

• Diagnostic and activity lights

• A power supply

Ethernet hubs and switches are made in two forms: managed and unmanaged. Managed hubs and switches can be configured, enabled or disabled, or monitored by a network operator and are commonly used on corporate networks. Workgroup and home-office networks use less expensive unmanaged hubs, which simply connect computers on the network.

Note The now-obsolete ARCnet network used its own types of hubs: passive hubs, which were unpowered, and active hubs, which used a power supply. Neither type of hub is compatible with Ethernet.

The connection between each workstation and the hub or switch is the UTP cable running from the RJ-45 jack on the rear of the NIC to the RJ-45 jack on the rear of the hub or switch.

Signal lights on the front of the hub or switch indicate which connections are in use by computers; switches also indicate whether a full-duplex connection is in use. Multispeed hubs and switches also indicate which connection speed is in use on each port. Your hub or switch must have at least one RJ-45 UTP connector for each computer you want to connect to it. Figure 20.13 shows a typical five-port, 10/100 Ethernet hub suitable for small networks at home or in a small business.

How Hubs Work

A computer on an Ethernet network broadcasts (sends) a request for network information or programs from a specific computer through the cable to the hub, which broadcasts the request to all computers connected to it. When the destination computer receives the message, it sends the requested information back to the hub, which broadcasts it again to all computers, although only the requesting computer acts on the information. Thus, a hub acts similarly to a radio transmitter and receiver that sends a signal to all radios, but only the radios set for the correct station can send or receive the information.

How Switches Differ from Hubs

Switches are similar to hubs in that they connect computers on a UTP-based Ethernet network to each other and physically resemble hubs (see Figure 20.14). However, instead of broadcasting data to all computers on the network as hubs do, switches use a feature called address storing, which checks the destination for each data packet and sends it directly to the computer it's intended for. Thus, a switch can be compared to a telephone exchange, making direct connections between the originator of a call and the receiver.

Because switches establish a direct connection between the originating and receiving PC, they also provide the full bandwidth of the network to each port. Hubs, by contrast, must subdivide the network's bandwidth by the number of active connections on the network, meaning that bandwidth rises and falls depending on network activity.

For example, assume you have a four-station network workgroup using 10/100 NICs and a Fast Ethernet hub. The total bandwidth of the network is 100Mbps. However, if two stations are active, the effective bandwidth available to each station drops to 50Mbps (100Mbps divided by 2). If all four stations are active, the effective bandwidth drops to just 25Mbps (100Mbps divided by 4)! Add more active users, and the effective bandwidth continues to drop.

By replacing the hub with a switch, the effective bandwidth for each station remains 100Mbps because the switch doesn't broadcast data to all stations.

Most 10/100 NICs and Fast Ethernet or 10/100 switches also support full-duplex (simultaneous transmit and receive), enabling actual bandwidth to be double the nominal 100Mbps rating: 200Mbps. Table 20.5 summarizes the differences between the two devices.

Table 20.5. Ethernet Hub and Switch Comparison

Feature	Hub	Switch
Bandwidth	Divided by total number of ports in use	Dedicated to each port in use
Data Transmission Type	Broadcast to all connected computers	Direct connection between transmitting and receiving computers
Duplex Support	Half-duplex only (receive or transmit)	Half- or full-duplex when (receive and transmit) used with full-duplex NICs; full-duplex operation doubles the effective bandwidth of the network

As you can see, using a switch instead of a hub greatly increases the effective speed of a network, even if all other components remain the same.

Because of the improved performance of switches, I recommend them instead of hubs for networks of any size.

Additional Hub and Switch Features You Might Need

Although older hubs and switches run at only a single speed and have only a few RJ-45 connectors, it makes sense to upgrade to newer, more flexible equipment. Most recent hubs and switches have the following useful features, which are worth asking for:

• Dual-speed hubs/switches. If you are adding Fast Ethernet (100BASE-TX) clients to an existing 10BASE-T network, you need a dual-speed hub or switch to connect the various types of Ethernet together.

  Even if you are building a brand-new Fast Ethernet network, a dual-speed hub or switch is useful for occasionally hosting a "guest" PC that has only a standard 10BASE-T Ethernet NIC onboard. Even though most Fast Ethernet switches and hubs on the market today are actually 10/100 dual-speed models, you might still encounter a Fast Ethernet-only unit. These should be used only on networks that will never have a 10BASE-T connection.

• Stackable hub or switch with an uplink port. A stackable hub or switch is one that can be connected to another hub, enabling you to add computers to your network without replacing the hub or switch every time it runs out of connections. Most hubs and switches on the market today are stackable (but very small) hubs, and some older models might lack this feature. You can use this feature to add 10/100 features to an older 10BASE-T-only network and connect a dual-speed hub or switch to the uplink port on your 10BASE-T hub.

• "Extra" ports beyond your current requirements. If you are connecting four computers together into a small network, you need a four-port hub or switch (the smallest available). But, if you buy a hub or switch with only four ports and want to add another client PC to the network, you must add a second hub or switch or replace the hub or switch with a larger one with more ports.

  Instead, plan for the future by buying a hub or switch that can handle your projected network growth over the next year. If you plan to add two workstations, buy at least a six-port hub or switch (the cost per connection drops as you buy hubs with more connections). Even though most hubs and switches are stackable to support the growth of your network, the more ports a hub or switch has, the less expensive per port it will be.

To determine whether a hub or switch is stackable, look for an uplink port, as shown in Figure 20.15. This port looks like an ordinary RJ-45 UTP port, but it is wired differently, enabling you to use a standard-pinout RJ-45 UTP cable to connect it to another hub. Without the uplink port, you'd have to use a specially wired crossover cable.

Typically, hubs and switches with an uplink port allow you to use the port along with all but one of the normal ports on the hub (see Figure 20.15). For example, one of my associates uses a five-port switch from Linksys that also contains a router (for Internet access) and an uplink port. If his office network expands beyond five computers, he can use the uplink port to add another switch to expand the network and provide the new stations, as well as the original network, with Internet access.

Hub and Switch Placement

Although large networks have a wiring closet near the server, the workgroup-size LANs you'll be building don't need a wiring closet. However, the location of the hub or switch is important.

Ethernet hubs and switches require electrical power, whether they are small units that use a power "brick" or larger units that have an internal power supply and a standard three-prong AC cord.

In addition to electrical power, consider placing the hub or switch where its signal lights will be easy to view for diagnostic purposes and where its RJ-45 connectors can be reached easily when it's time to add another user or two. In many offices, the hub or switch sits on the corner of the desk, enabling the user to see network problems just by looking at the hub or switch.

If the hub or switch also integrates a router for use with a broadband Internet device, such as a DSL or cable modem, you can place it near the cable or DSL modem or at a distance if the layout of your home or office requires it. Because the cable or DSL modem usually connects to your computer by the same Category 5 cable used for UTP Ethernet networking, you can run the cable from the cable or DSL modem to the router/switch's WAN port and connect all the computers to the LAN ports on the router/switch.

Except for the 328" (100 meter) limit for all forms of UTP Ethernet (10BASE-T, 100BASE-TX, and 1000BASE-TX), distances between each computer on the network and the hub or switch aren't critical, so put the hub or switch wherever you can supply power and gain easy access.

Tip Decide where you plan to put your hub or switch before you buy prebuilt UTP wiring or make your own; if you move the hub or switch, some of your wiring will no longer be the correct length. Although excess lengths of UTP cable can be coiled and secured with cable ties, cables that are too short should be replaced. You can buy RJ-45 connectors to create one long cable from two short cables, but you must ensure the connectors are Category 5 if you are running Fast Ethernet; some vendors still sell Category 3 connectors that support only 10Mbps. You're really better off replacing the too-short cable with the correct length.