This chapter is from the book
This chapter is from the book
This chapter is from the book
Choosing LAN Cabling Options
Several types of cables and connectors can be used in LANs, depending on the require-ments for the network and the type of Ethernet to be implemented. These connectors also vary depending on the type of media that you have installed.
Learning about the different types of cables and connectors in an Ethernet LAN and their various functions can help you understand more about how a LAN works.
LAN Physical Layer
Ethernet is the most widely used LAN technology. Since its initial implementation, Ethernet has been extended to four new types:
802.3u (Fast Ethernet)
802.3z (Gigabit Ethernet over Fiber)
802.3ab (Gigabit Ethernet over UTP)
802.3ae (10 Gigabit Ethernet)
The cabling aspects of the LAN exist at Layer 1 of the Open System Interconnection (OSI) reference model. Figure 4-7 shows a subset of physical layer implementations that can be deployed to support Ethernet.
)
Figure
4-7 Ethernet at the Physical Layer
Ethernet in the Campus
Before implementing a network, you need to determine the requirements for the network. You can remember a few common recommendations on how various Ethernet technologies can be used in a campus network environment.
In many modern installations, infrastructure costs for cabling and adapters can be high. Using the appropriate Ethernet connectivity provides the necessary speed for the parts of the network that require it while controlling costs.
In general, you can use Ethernet technologies in a campus network in several different ways:
An Ethernet speed of 10 Mbps can be used at the access layer to provide adequate performance for most users. In addition, 100-Mbps Fast Ethernet can be used for high-bandwidth-consuming clients or servers.
Gigabit Ethernet is typically used as the link between the access layer and network devices, supporting the aggregate traffic from each Ethernet segment on the access link.
To enhance client-server performance across the campus network and avoid bottlenecks at the server, Fast Ethernet or Gigabit Ethernet links can be used to connect enterprise servers. Gigabit Ethernet, in combination with switched Fast Ethernet, creates an effective solution for avoiding slow networks.
Gigabit Ethernet links can provide the connection between the distribution layer and the core. Because the campus network model supports dual links between each distribution layer router and core switch, you can load balance the aggregate traffic from multiple-access switches across the links.
Gigabit Ethernet (or 10 Gigabit Ethernet) should be used between switches and the backbone. The fastest affordable media should be implemented between backbone switches.
Table 4-2 outlines the recommendations for Ethernet deployment.
Table 4-2 Ethernet Connectivity Recommendations
|
Network Hierarchy Layer |
Ethernet 10 Mbps |
Fast Ethernet 100 Mbps |
Gigabit Ethernet 1000 Mbps |
10 Gigabit Ethernet 10000 Mbps |
|
Access layer |
Connects users with low to moderate bandwidth requirements |
Connects users with high-speed requirements or servers with low to moderate usage |
Connects servers with high usage |
Not currently recommended at this layer |
|
Distribution layer |
Not recommended at this layer |
Connects routers and switches with moderate usage |
Interconnects access switches with Fast Ethernet users and used to connect distribution switches to core layer |
Not currently recommended at this layer |
|
Core layer |
Not recommended at this layer |
Not recommended at this layer |
Interconnects core switches in networks with moderate use |
Interconnects core switches with high usage |
NOTE
Currently, some organizations are considering providing Gigabit Ethernet to the end user; however, not many applications can take full advantage of this infrastructure, and providing Gigabit Ethernet to the end user can potentially create a bottleneck between network devices. You should consider this carefully before installing gigabit technology to the end users.
Ethernet Media and Connector Requirements
In addition to considering the requirements for the Ethernet LAN, the media and connector requirements for each implementation must be considered. This topic outlines the cable and connector specifications used to support Ethernet implementations.
The cable and connector specifications used to support Ethernet implementations are derived from the Electronic Industries Alliance and (newer) Telecommunications Industry Alliance (EIA/TIA) standards body. The categories of cabling defined for Ethernet are derived from the EIA/TIA-568 (SP-2840) Commercial Building Telecommunications Wiring Standards. EIA/TIA specifies an RJ-45 connector for UTP cable. The letters RJ stand for registered jack, and the number 45 refers to a specific physical connector that has eight conductors.
Table 4-3 compares the cable and connector specifications for the most popular Ethernet implementations.
The important difference to note is the media used for 10-Mbps Ethernet versus 100-Mbps Ethernet. In today's networks, in which you see a mix of 10- and 100-Mbps requirements, you must be aware of the need to change over to UTP Category 5 to support Fast Ethernet.
Connection Media
Several connection media can be used in an Ethernet LAN implementation. Figure 4-8 illustrates different connection types—attachment unit interface (AUI), RJ-45, and gigabit—used by each physical layer implementation. The RJ-45 connector and jack are the most prevalent. RJ-45 connectors are discussed in more detail later in this chapter.
In some cases, the type of connector on a NIC does not match the type of media that it needs to connect to. As shown in Figure 4-8, an interface exists for the AUI connector on many Cisco devices. The AUI is the 15-pin physical connector interface between a computer's NIC and coaxial Ethernet cable.
Table 4-3 TCable and Connector Specifications
|
|
10BASE-2 |
10BASE-5 |
10BASE-T |
100BASE-TX |
100BASE- FX |
1000 BASE-CX |
1000 BASE-T |
1000 BASE-SX |
1000 BASE-LX |
|
Media |
50-ohm coaxial (Thinnet) RG-58 coaxial cable |
50-ohm coaxial (Thicknet) RG-50 coaxial cable |
EIA/TIA Category 3, 4, 5 UTP 2 pair |
EIA/TIA Category 5 UTP 2 pair |
62.5/125 micro multimode fiber |
STP |
EIA/TIA Category 5 UTP 4 pair |
62.5/50 micro multimode fiber |
9 micro single-mode fiber |
|
Maximum Segment Length |
185 m (606.94 ft) |
500 m (1640.4 ft) |
100 m (328 ft) |
100 m (328 ft) |
400 m (1312.3 ft) |
25 m (82 ft) |
100 m (328 ft) |
260 m (853 ft) |
3-10 km (1.86–6.2 miles) |
|
Topology |
Bus |
Bus |
Star |
Star |
Point-to-point |
Star or point-to-point |
Star or point-to-point |
Point-to-point |
Point-to-point |
|
Connector |
AUI or BNC connector |
AUI |
ISO 8877 (RJ-45) |
ISO 8877 (RJ-45) |
MT-RJ or SC connector |
ISO 8877 (RJ-45) |
ISO 8877 (RJ-45) |
SC |
SC |
A Gigabit Interface Converter (GBIC), like the one shown in Figure 4-9, is a hot-swappable input/output device that plugs into a Gigabit Ethernet port. A key benefit of using a GBIC is that GBICs are interchangeable. This allows users the flexibility to deploy other 1000BASE-X technology without needing to change the physical interface/model on the router or switch. GBICs support UTP (copper) and fiber-optic media for Gigabit Ethernet transmission.
)
Figure
4-9 GBIC
<Anchor0>
Typically, GBICs are used in the LAN for aggregation and in the backbone. You also see GBICs in SANs and MANs.
The fiber-optic GBIC is a transceiver that converts serial electric currents to optical signals and optical signals to digital electric currents. Some of the optical GBICs include the following:
Short wavelength (1000BASE-SX)
Long wavelength/long haul (1000BASE-LX/LH)
Extended distance (1000BASE-ZX)
UTP Implementation
In a UTP implementation, you must determine the EIA/TIA type of cable and whether to use a straight-through or crossover cable. This section describes the types of connectors used in a UTP implementation and the characteristics and uses of straight-through and crossover cables.
If you look at an RJ-45 transparent end connector, like the one in Figure 4-10, you can see eight colored wires, twisted into four pairs. Four of the wires (two pairs) carry the positive or true voltage and are considered tip (T1 through T4); the other four wires carry the inverse of false voltage grounded and are called ring (R1 through R4). Tip and ring are terms that originated in the early days of the telephone. Today, these terms refer to the positive and the negative wires in a pair. The wires in the first pair in a cable or a connector are designated as T1 and R1, the second pair is T2 and R2, and so on.
)
Figure
4-10 RJ-45 Connector
The RJ-45 plug is the male component, crimped at the end of the cable. As you look at the male connector from the front (the side with the metal pins exposed), the pin locations are numbered from 8 on the left to 1 on the right.
The RJ-45 jack, shown in Figure 4-11, is the female component in a network device, wall, cubicle partition outlet, or patch panel.
)
Figure
4-11 RJ-45 Jack
In addition to identifying the correct EIA/TIA category of cable to use for a connecting device (depending on what standard is being used by the jack on the network device), you need to determine which of the following to use:
A straight-through cable
A crossover cable
The RJ-45 connectors on both ends show all the wires in the same order. If the two RJ-45 ends of a cable are held side by side in the same orientation, the colored wires (or strips or pins) are seen at each connector end. If the order of the colored wires is the same at each end, the cable is straight-through. Figure 4-12 shows the wiring for a straight-through cable.
)
Figure
4-12 Straight-Through Cable Wiring
With crossover, the RJ-45 connectors on both ends show that some of the wires on one side of the cable are crossed to a different pin on the other side of the cable. Specifically, for Ethernet, pin 1 at one RJ-45 end should be connected to pin 3 at the other end. Pin 2 at one end should be connected to pin 6 at the other end, as shown in Figure 4-13.
)
Figure
4-13 Crossover Cable Wiring
Each device using RJ-45 connectors transmits or receives on particular pins depending on the device type. A PC or router typically transmits on pins 1 and 2 while a switch or hub receives on pins 1 and 2. You must follow certain guidelines when connecting these devices.
Use straight-through cables for the following cabling:
Switch to router
Switch to PC or server
Hub to PC or server
Use crossover cables for the following cabling:
Switch to switch
Switch to hub
Hub to hub
Router to router
PC to PC
Router to PC
Occasionally, ports on network devices are marked with an X, like those in Figure 4-14. This marking means that these devices receive on pins 1 and 2, or that they are crossed. When connecting devices in a network, you might be required to use a variety of cable types.
NOTE
Most hubs and some switches have a port that can be changed from X to not-X by moving a switch or pressing a button. This feature enables you to use straight-through cables where crossovers would ordinarily be required.
)
Figure
4-14 RJ-45 Port Designations
Being aware of the different cabling media types, specifications, and connectors is an important step to interconnecting network devices.