Wire Strippers

What do you want to strip today? The variety of cable strippers represented in this section is a function of the many different types of cable you can work with, different costs of the cable strippers, and versatility of the tools. Strippers for UTP, ScTP, and STP cables are used to remove the outer jacket and have to accommodate the wide variation in the geometry of UTP cables. Unlike coax, which is usually

consistently smooth and round, twisted-pair cables can have irregular surfaces due to the jacket shrinking down around the pairs. Additionally, the jacket thickness can differ greatly depending on brand and flame rating. The trick is to aid removal of the jacket without nicking or otherwise damaging the insulation on the conductors underneath. The wire stripper in Figure 6.1 uses an adjustable blade so that you can fix the depth, matching it to the brand of cable you are working with. Some types use spring tension to help keep the blade at the proper cutting depth.

In both cases, the goal is to score (lightly cut) the jacket without penetrating it completely. Then, you flex the cable to break the jacket along the scored line. This ensures that the wire insulation is nick-free. In some models, the tool can also be used to score or slit the jacket lengthwise in the event you need to expose a significant length of conductors.

Modular Patch Cables

Modular patch cables (patch cords) are used to provide the connection between field-terminated horizontal cables and network-connectivity devices such as switches and hubs and connections between the wall-plate jack and network devices such as computers. They are the part of the network wiring you can actually see. As the saying goes, a chain is only as strong as its weakest  link. Because of their exposed position in structured cable infrastructures, modular patch cords are almost always the weakest link. Whereas horizontal UTP cables contain solid conductors, patch cords are made with

stranded conductors because they are more flexible. The flexibility allows them to withstand the abuse of frequent flexing and reconnecting. Although you could build your own field terminated patch cords, we strongly recommend against it. The manufacture of patch cords is very exacting, and even under controlled factory conditions it is difficult to achieve and guarantee consistent transmission performance. The first challenge lies within the modular plugs themselves. The parallel alignment of the contact blades forms a capacitive plate, which becomes a source of signal coupling or crosstalk. Further, the untwisting and splitting of the pairs as a result of the termination process increases the cable's susceptibility to crosstalk interference. If that weren't enough, the mechanical crimping process that secures the plug to the cable could potentially disturb the cable's normal geometry by crushing the conductor pairs. This is yet another source of crosstalk interference and a source of attenuation.

Routers

Routers are packet-forwarding devices just like switches and bridges; however, routers allow transmission of data between network segments. Unlike switches, which forward packets based on physical node addresses, routers operate at the network layer of the OSI reference model, forwarding packets based on a network ID.If you recall from our communication digression in the discussion on bridging, we defined a network as a logical grouping of computers and network devices. A collection of interconnected networks is referred to as an internetwork. Routers provide the connectivity within an internetwork. So how do routers work? In the case of the IP protocol, an IP address is 32 bits long. Those 32 bits contain both the network ID and the host ID of a network device. IP distinguishes betweennetwork and host bits by using a subnet mask. The subnet mask is a set of contiguous bits with values of one from left to right, which IP considers to be the address of a network. Bits used to describe a host are masked out by a value of 0, through a binary calculation process called AND ing. Figure shows two examples of network IDs calculated from an ANDing process. We use IP as the basis of our examples because it is the industry standard for enterprise networking; however, TCP/IP is not the only routable protocol suite. Novell's IPX/SPX and Apple Computer's AppleTalk protocols are also routable.



Switches

A switch is the next rung up the evolutionary ladder from bridges. In modern star-topology networking, when you need bridging functionality you often buy a switch. But bridging is not the only benefit of switch implementation. Switches also provide the benefit of micro-LAN segmentation, which means that every node connected to a switched port receives its own dedicated bandwidth. And with switching, you can further segment the network into virtual LANs. Like bridges, switches also operate at the link layers of the OSI reference model and, in the case of Layer-3 switches, extend into the network layer. The same mechanisms are used to build dynamic tables that associate MAC addresses with switched ports. However, whereas bridges implement store-and-forward bridging via software, switches implement either store and-forward or cut-through switching via hardware, with a marked improvement of speed. Micro-LAN segmentation is the key benefit of switches, and most organizations have either completely phased out hubs or are in the process of doing so to accommodate the throughput requirements for multimedia applications. Although switches are becoming more affordable, ranging in price from $10 to slightly over $20 per port, their price may still prevent organizations from migrating to completely switched infrastructures. At a minimum, however, servers and workgroups should be linked through switched ports.

Bridges

When we use the terms bridge and bridging, we are generally describing functionality provided by modern switches. Just like a repeater, a bridge is a network device used to connect two network segments. The main difference between them is that bridges operate at the link layer of the OSI reference model and can therefore provide translation services required to connect dissimilar media access architectures such as Ethernet and Token Ring. Therefore, bridging is an important internetworking technology.

In general, there are four types of bridging:

Transparent bridging Typically found in Ethernet environments, the transparent bridge

analyzes the incoming frames and forwards them to the appropriate segments one hop at a time

Source-route bridging Typically found in Token Ring environments, source-route

bridging provides an alternative to transparent bridging for NetBIOS and SNA protocols. In source route bridging, each ring is assigned a unique number on a source-route bridge port. Token Ring frames contain address information, including a ring and bridge numbers, which each bridge analyzes to forward the frame to the appropriate ring

Hubs

Because repetition of signals is a function of repeating hubs, hub and repeater are used interchangeablywhen referring to twisted-pair networking. The semantic distinction between the two terms is that a repeater joins two backbone coaxial cables, whereas a hub joins two or more twisted-pair cables. In twisted-pair networking, each network device is connected to an individual network cable. In coaxial networking, all network devices are connected to the same coaxial backbone. A hub eliminates the need for BNC connectors and vampire taps. Figure illustrates how network devices connect to a hub versus to coaxial backbones. Hubs work the same way as repeaters in that incoming signals are regenerated before they are retransmitted across its ports. Like repeaters, hubs operate at the OSI physical layer, which means they do not alter or look at the contents of a frame traveling across the wire. When a hub receives an incoming signal, it regenerates it and sends it out over all its ports. Figure shows a hub at work.

Repeaters

Nowadays, the terms repeater and hub are used synonymously, but they are actually not the same. Prior to the days of twisted-pair networking, network backbones carried data across coaxial cable, similar to what is used for cable television. Computers would connect into these either by BNC connectors, in the case of thinnet, or by vampire taps, in the case of thicknet. Everyone would be connected to the same coaxial back-bone. Unfortunately, when it comes to electrical current flowing through a solid medium, you have to contend with the laws of physics. A finite distance exists in which electrical signals can travel across a wire before they become too distorted. Repeaters were used with coaxial cable to overcome this challenge.

Repeaters work at the physical layer of the OSI reference model. Digital signals decay due to attenuation and noise. A repeater's job is to regenerate the digital signal and send it along in its original state so that it can travel farther across a wire. Figure illustrates a repeater in action.

Asynchronous Transfer Mode (ATM)

ATM (asynchronous transfer mode, not to be confused with automated teller machines) first emerged in the early 1990s. If networking has an equivalent to rocket science, then ATM is it. ATM was designed to be a high-speed communications protocol that does not depend on any specific LAN topology. It uses a high-speed cell-switching technology that can handle data as well as real-time voice and video. The ATM protocol breaks up transmitted data into 48- byte cells that are combined with a 5-byte header. A cell is analogous to a data packet or frame. 

ATM is designed to "switch" these small, fixed-size cells through an ATM network very quickly. It does this by setting up a virtual connection between the source and destination nodes; the cells may go through multiple switching points before ultimately arriving at their final destination. If the cells arrive out of order, and if the implementation of the receiving system is set up to do so, the receiving system may have to correctly order the arriving cells. ATM is a connection-oriented service, in contrast to many network architectures, which are broadcast based. Connection orientation simply means that the existence of the opposite end is established through manual setup or automated control information before user data is transmitted. Data rates are scalable and start as low as 1.5Mbps, with other speeds of 25-, 51-, 100-, and 155Mbps and higher. The most common speeds of ATM networks today are 51.84Mbps and 155.52Mbps. Both of these speeds can be used over either copper or fiber-optic cabling. A 622.08Mbps ATM is also becoming common but is currently used exclusively over fiber-

optic cable, mostly as a network backbone architecture. ATM supports very high speeds because it is designed to be implemented by hardware rather than software and is in use at speeds as high as 10Gbps. In the United States, the specification for synchronous data transmission on optical media is

SONET (Synchronous Optical Network); the international equivalent of SONET is SDH (Synchronous Digital Hierarchy). SONET defines a base data rate of 51.84Mbps; multiples of this rate are known as optical carrier (OC) levels, such as OC-3, OC-12, etc.

Fiber Distributed Data Interface (FDDI)

Fiber Distributed Data Interface (FDDI) is a networking specification that was produced by the ANSI X3T9.5 committee in 1986. It defines a high-speed (100Mbps), token-passing network using fiber-optic cable. In 1994, the specification was updated to include copper cable. The copper cable implementation was designated TP-PMD, which stands for Twisted Pair-Physical Media Dependent. FDDI was slow to be widely adopted, but for awhile it found a niche as a reliable, high-speed technology for backbones and applications that demanded reliable connectivity.

Though at first glance FDDI appears to be similar to Token Ring, it is different from both Token Ring and Ethernet. A Token Ring node can transmit only a single frame when it gets the free token; it must wait for the token to transmit again. An FDDI node, once it possesses the free token, can transmit as many frames as it can generate within a predetermined time before it has to give up the free token.

FDDI can operate as a true ring topology, or it can be physically wired like a star topology. Figure shows an FDDI ring that consists of dual-attached stations (DAS); this is a true ring topology. A dual attached station has two FDDI interfaces, designated as an A port and a B port. The A port is used as a receiver for the primary ring and as a transmitter for the secondary ring. The B port does the opposite: it is a transmitter for the primary ring and a receiver for the secondary ring. Each node on the network in Figure 3.9 has an FDDI network-interface card that has two FDDI attachments. The card creates both the primary and secondary rings. Cabling for such a network is a royal pain because the cables have to form a complete circle.

Token Ring

Developed by IBM, Token Ring uses a ring architecture to pass data from one computer to another. A former teacher of Jim's referred to Token Ring as the Fahrenheit network architecture because more people with Ph.D. degrees worked on it than there are degrees in the Fahrenheit scale. Token Ring employs a sophisticated scheme to control the flow of data. If no network node needs to transmit data, a small packet, called the free token, continually circles the ring. If a node needs to transmit data, it must have possession of the free token before it can create a new Token Ring data frame. The token, along with the data frame, is sent along as a busy token. Once the data arrives at its destination, it is modified to acknowledge receipt and sent along again until it arrives back at the original sending node. If there are no problems with the correct receipt of the packet, the original sending node releases the free token to circle the network again. Then another node on the ring can transmit data if necessary.

Gigabit Ethernet (1000Mbps)

The IEEE approved the first Gigabit Ethernet specification in June 1998-IEEE 802.3z. The purpose of IEEE 802.3z was to enhance the existing 802.3 specification to include 1000Mbps operation (802.3 supported 10Mbps and 100Mbps). The new specification covers media access control, topology rules, and the gigabit media-independent interface. IEEE 802.3z specifies three physical layer interfaces: 1000Base-SX, 1000Base-LX, and 1000Base-CX. In July 1999, the IEEE approved an additional specification known as IEEE 802.3ab, which adds an additional Gigabit Ethernet physical layer for 1000Mbps over UTP cabling. The UTPcabling, all components, and installation practices must be Category 5 or greater. The only caveat is that legacy (or new) Category 5 installations must meet the performance requirementsoutlined in ANSI/TIA/EIA-568-B.

Gigabit Ethernet deployment is still in the early stages, and we don't expect to see it extended directly to the desktop in most organizations. The cost of Gigabit Ethernet hubs and network- interface cards is too high to permit this in most environments. Only applications that demand the highest performance will actually see Gigabit Ethernet to the desktop in the next few years. Initially, the most common uses for Gigabit Ethernet will be for intrabuilding or campus backbones. Figure shows a before-and-after illustration of a simple network with Gigabit Ethernet deployed. Prior to deployment, the network had a single 100Mbps switch as a back- bone for several 10Mbps and 100Mbps segments. All servers were connected to the 100Mbpsbackbone switch, which was sometimes a bottleneck.

100Base-FX Ethernet

Like its 100Base-TX copper cousin, 100Base-FX uses a physical-media specification developed by ANSI for FDDI. The 100Base-FX specification was developed to allow 100Mbps Ethernet to be used over fiber-optic cable. Though the cabling plant is wired in a star topology, 100Base-FX is a bus architecture. If you choose to use 100Base-FX Ethernet, consider the following: Cabling-plant topology should be a star topology and should follow ANSI/TIA/EIA-568-B or ISO 11801 recommendations.

Each network node location should have a minimum of two strands of multimode fiber (MMF). Maximum link distance is 400 meters; though fiber-optic cable can transmit over much farther distances, proper signal timing cannot be guaranteed. If you follow ANSI/TIA/EIA-568-B or ISO 11801 recommendations, the maximum horizontal-cable distance should not exceed 100 meters.

The most common fiber connector type used for 100Base-FX is the SC connector, but the ST connector and the FDDI MIC connector may also be used. Make sure you know which type of connector(s) your hardware vendor will require.

100Base-TX Ethernet

The 100Base-TX specification uses physical-media specifications developed by ANSI that were originally defined for FDDI (ANSI specification X3T9.5) and adapted for twisted-pair cabling. The 100Base-TX requires Category 5 or better cabling but uses only two of the four pairs. The eight-position modular jack (RJ-45) uses the same pin numbers as 10Base-T Ethernet. Though a typical installation requires hubs or switches, two 100Base-TX nodes can be connected together "back-to-back" with a crossover cable made exactly the same way as a 10Base-T crossover cable. (See Chapter 9, "Connectors," for more information on making a 10Base-T or 100Base-TX crossover cable.) Understand the following when planning a 100Base-TX Fast Ethernet network:
All components must be Category 5 or better certified, including cables, patch panels, and connectors. Proper installation practices must be followed. If you have a Category 5 "legacy" installation, the cabling system must be able to pass tests specified by Annex N of ANSI/TIA/EIA-568-B.2. The maximum segment cable length is 100 meters. With higher-grade cables, longer lengths of cable may work, but proper signal timing cannot be guaranteed.