Table of Contents

Installation Preparation

Installation Preparation

This chapter describes how to prepare your site before installing switching modules in Catalyst 5000 family switches and consists of these sections:

• Site Considerations
  
  
• Determining Interface Connector and Cable Requirements
  
  

Warning Before you install, operate, or service the system, read the Site Preparation and Safety Guide. This guide contains important safety information you should know before working with the system.

Site Considerations

This section discusses several topics you should consider before installing switching modules in Catalyst 5000 family switches.

Limiting Connection Distances

The length of your networks and the distances between connections depend on the signal type, the signal speed, and the transmission media (the type of cabling used to transmit the signals). For example, fiber-optic cable has a greater channel capacity than twisted-pair cabling. The distance and rate limits in this chapter are the IEEE-recommended maximum speeds and distances for signaling. However, if you understand the electrical problems that might arise and can compensate for them, you should get good results with rates and distances greater than those described here, although you do so at your own risk.

Preparing Network Connections

When preparing your site for network connections to the switching modules, you need to consider factors for each interface type:

• Type of cabling (fiber, thick, or twisted-pair cabling)
  
  
• Distance limitations
  
  
• Specific cables to connect each interface
  
  
• Any additional interface equipment, such as transceivers, modems, channel service units (CSUs), or data service units (DSUs)
  
  

Before installing the switching modules, have all additional external equipment and cables on hand. If you intend to build your own cables, see Appendix A, "Specifications."

Optional Connection Equipment

You might need some of the following data communications equipment to complete your installation:

• Ethernet transceiver and transceiver cable, if you plan to use an IEEE 802.3 or Ethernet interface (thick-wire or thin-wire) at your installation.
  
  
• An optical bypass switch, if you plan to use the optical bypass feature available with single-mode and multimode Fiber Distributed Data Interface (FDDI) interfaces.
  
  

Determining Interface Connector and Cable Requirements

This section describes the types of cables and connectors you will need to install the switching modules.

Cabling Considerations for Ethernet and Fast Ethernet Modules

Ethernet and Fast Ethernet modules use six different types of connectors:

• 40-pin media-independent interface (MII) male connectors to connect Fast Ethernet external transceivers (see Figure 2-1)
  
• RJ-21 telco data bus (DB-21) male connectors to connect Ethernet or Fast Ethernet connections to a punch-down block or a patch panel (see Figure 2-2 and Figure 2-3)
  
• RJ-45 male connectors to connect Ethernet and Fast Ethernet (see Figure 2-4)
  
• Straight-tip (ST)-type fiber-optic connectors to connect Ethernet (see Figure 2-5)
  
• Simplex-connector (SC)-type fiber-optic connectors to connect Fast Ethernet connections (see Figure 2-6)
  
• MT-RJ fiber-optic connectors to connect Ethernet connections (see Figure 2-7)
  

The 24-port, switched 10BaseT module can use both 90-degree and 180-degree male telco connectors, as shown in Figure 2-2 and Figure 2-3. The 48-port 10BaseT modules require 180-degree male telco connectors.

Ethernet transceivers for Virtual Terminal Protocol (VTP) and multimode fiber-optic cabling (100BaseFX at 100 Mbps) are available from a variety of suppliers.

Figure 2-8 shows examples of Fast Ethernet transceivers and connection equipment.

When planning your connections, consider the types and locations of connectors on adjacent switching modules so that the transceiver does not overlap and impair access to other connections.

The maximum transmission distances for Ethernet network segments and connections depend on the type of transmission cable used; for example, unshielded twisted-pair (UTP).

Table 2-1 lists the IEEE maximum transmission distances for Ethernet and Fast Ethernet.

Cabling Considerations for Gigabit Ethernet and Gigabit EtherChannel Modules

Both the Gigabit Ethernet module (WS-X5403) and the Gigabit EtherChannel module (WS-X5410) use Gigabit Interface Converters (GBICs). The GBIC is a hot-swappable input/output device that plugs into a Gigabit Ethernet or EtherChannel module, linking the module with the fiber-optic network. All GBIC ports have SC-type connectors. A GBIC is shown in Figure 2-9.

Table 2-2 lists the three styles of GBICs that are currently available.

Table 2-3 provides cabling specifications for the GBICs that you install in the Gigabit Ethernet or Gigabit EtherChannel module.

GBIC Optical Power Characteristics

Table 2-4 provides the GBIC optical power characteristics.

GBIC Cabling Restrictions

You must observe the following optical-fiber cabling restrictions when using GBICs:

• The minimum cabling distance for 1000BaseSX and 1000BaseLX/LH GBICs is 6.5 feet (2 meters).
  
  
• When using the 1000BaseLX/LH GBIC with 62.5-micron diameter MMF, you must install a mode-conditioning patch cord between the MMF fiber-optic network and the GBIC. The mode-conditioning patch cord (CAB-GELX-625 or equivalent) is required to comply with IEEE standards. The IEEE found that link distances could not be met with certain types of fiber-optic cable cores. The solution is to launch light from the laser at a precise offset from the center by using the mode-conditioning patch cord. At the output of the patch cord, the LX/LH GBIC is compliant with the IEEE 802.3z standard for 1000BaseLX.
  
  
• You must insert a 10-dB inline optical attenuator between the single-mode fiber-optic network and the receiving port on the 1000BaseZX GBIC at each end of the link if the link length is less than 15.5 miles (25 km).
  
  
• You must insert a 5-dB inline optical attenuator between the single-mode fiber-optic network and the receiving port on the 1000BaseZX GBIC at each end of the link if the link is greater than 15.5 miles (25 km), but less than 31 miles (50 km).
  
  

Cabling Considerations for FDDI and CDDI Modules

CDDI is the implementation of FDDI protocols over foil twisted-pair (FTP) and UTP cabling. CDDI transceivers transmit over short distances (about 100 feet), providing data rates of 100 Mbps using a dual-ring architecture to provide redundancy.

FDDI and CDDI standards set the maximum allowable distances between stations and the maximum allowable cable lengths. Table 2-5 lists the maximum transmission distances for each transceiver type.

Table 2-6 lists the typical fiber-optic link attenuation and dispersion limits.

The standards allow a maximum of 500 stations in an FDDI/CDDI network. Both single-mode and multimode transceiver types provide 11 dB of optical power.

Setting Up FDDI/CDDI Connection Equipment

Fiber-optic transceivers on the FDDI/CDDI modules provide a direct interface between the switch and the FDDI/CDDI ring. The FDDI modules support multimode transceivers. Multimode transceivers provide a Class A dual-attachment interface that can connect to a Class A or a Class B station. Class A is a dual-attachment station (DAS) with primary and secondary rings; Class B is a single-attachment station (SAS) with only a primary ring. See the "Setting Up the FDDI/CDDI Network" section, for a detailed description of Class A and Class B stations and of DASs and SASs.

Using FDDI Media

FDDI networks use two types of fiber-optic cable: single-mode fiber (SMF) and multimode fiber (MMF). The term mode refers to the angle that light rays (signals) reflect and propagate through the optical-fiber core, which acts as a waveguide for the light signals. MMF has a thick core (62.5 micron) that reflects light rays at many angles. SMF has a thin core (8.7 to 10 micron) that allows light to enter only at a single angle.

Although MMF allows more light signals to enter at a greater variety of angles (modes), the different angles create multiple propagation paths that cause the signals to spread out in time and limit the rate at which data can be accurately received. This distortion does not occur on the single path of the single-mode signal; therefore, SMF is capable of higher bandwidth and greater cable run distances than MMF. Typically, multimode transmitters use LEDs as a light source. Single-mode transmitters use a laser diode, which can sustain faster data rates. Single-mode and multimode interfaces use a photodiode detector at the receiver to translate the light signal into electrical signals. Table 2-7 lists the FDDI transmitter power and receiver sensitivities for MMF and SMF.

Table 2-8 shows the transmit power levels for FDDI media.

Table 2-9 shows the receive power levels for FDDI media.

Configuring CDDI Transceivers and Cable Connections

The CDDI transceiver supports distances of up to 328 feet (100 meters). The CDDI connector is a CDDI-standard physical sublayer (PHY) connector that encodes and decodes the data into a format acceptable for UTP transmission. The CDDI connector accepts RJ-45 connectors, shown in Figure 2-10, with standard UTP cables.

Before you install CDDI modules, confirm that all existing cables conform with CDDI distance requirements and ensure that you have the proper modular RJ-45 connectors. CDDI cable and distance specifications are as follows:

• Cable—Data-grade UTP wiring; EIA/TIA-568, Category 5, data-grade cable is required for CDDI installations.
  

• Distance—The total length of data-grade UTP cable from one switch to another switch, station, or CDDI concentrator must not exceed 328 feet (100 meters), including patch cords and cross-connect jumpers.
  

When you plan your CDDI installation, remember the following:

• Use cross-connect (patch) panels that comply with the EIA/TIA-568, Category 5 wiring standard.
  
  
• Do not use bridge taps.
  
• Do not use protection coils.
  
• Do not share services (such as voice and data on the same cable). CDDI uses two of the four pairs in the twisted-pair cable. You cannot use the remaining two pairs for other applications.
  
• Do not exceed the maximum cable length for CDDI FTP and UTP of 328 feet (100 meters).
  

Configuring FDDI Transceivers and Cable Connections

The multimode transceiver supports distances of up to 1.2 miles (2 kilometers). The multimode connector is an FDDI-standard PHY connector that encodes and decodes the data into an acceptable format for fiber transmission. The multimode connector accepts standard 62.5/125-micron, multimode fiber-optic cable and with proper cable terminators, the connector can accept 50/125-micron fiber-optic cable. Multimode fiber-optic cable uses the integrated media interface connector (MIC), shown in Figure 2-11, at the FDDI module and the network ends.

The single-mode transceivers support distances up to 18.6 miles (30 km). The single-mode connector, shown in Figure 2-12, accepts standard 8.7- or 10/125-micron, single-mode fiber-optic cable using ST-type connectors for transmit and receive ports.

The FDDI modules provide a control port for an optical bypass switch. The control port allows the light signal to pass directly through the bypass switch and completely bypass the FDDI module transceivers when the interface fails or is shut down. Most optical bypass switches provide the necessary interface cables for connection to the MIC connectors on the FDDI module. However, not all manufacturers use the same type of DIN connector for the control port; some manufacturers use a DIN, and some use a smaller version, a mini-DIN.

Setting Up the FDDI/CDDI Network

FDDI/CDDI, which specifies a 100-Mbps, token-passing, dual-ring network using UTP or fiber-optic transmission media, is defined by the American National Standards Institute (ANSI) X3.1 standard and by ISO 9314, the international version of the ANSI standard.

An FDDI/CDDI network consists of two token-passing rings: a primary ring and a secondary ring. An FDDI/CDDI ring consists of two or more point-to-point connections between adjacent stations.

On most FDDI/CDDI networks, the primary ring is for data communication and the secondary ring is a backup. Class B stations, or SASs, attach to one ring and attach through a concentrator, which provides connections for multiple SASs. Class A stations, or DASs, attach to both rings.

Figure 2-13 shows a typical FDDI/CDDI configuration with a DAS, SAS, and concentrator.

The concentrator in a typical FDDI/CDDI configuration ensures that a failure or power-down of any SAS does not interrupt the ring. SASs (Class B) use one transmit port and one receive port to attach to the single ring. DASs (Class A) use two physical ports, designated PHY A and PHY B, each of which connects the station to both the primary and secondary rings. Each port is a receiver for one ring and a transmitter for the other. For example, PHY A receives traffic from the primary ring, and PHY B transmits to the primary ring.

The dual rings in an FDDI/CDDI network provide fault tolerance. If a station on a dual ring shuts down or fails, such as Station 3 in Figure 2-14, the ring wraps automatically (doubles back on itself) to form a single contiguous ring. This action removes the failed station from the ring, but allows the other stations to continue operation. In Figure 2-14, the ring wraps to eliminate Station 3 and forms a smaller ring that includes only Stations 1, 2, and 4.

A second failure could cause the ring to wrap in both directions from the point of failure, which would segment the ring into two separate rings that could not communicate with each other. For example, if Station 1 in Figure 2-14 fails after Station 3 fails, Stations 2 and 4 are isolated because no path for communication exists between them. Subsequent failures cause additional segmentation.

Each station in a ring refers to its neighbor stations as upstream or downstream neighbors. The stream is based on the signal flow on the primary ring. A station receives the primary signal from its upstream neighbor and transmits the primary signal to its downstream neighbor.

For example, Figure 2-14 shows the primary signal flow transmitted from PHY B on Station 2 to PHY A on Station 1 and from PHY B on Station 1 to PHY A on Station 4. Using Station 1 as a reference, Station 2 is the upstream neighbor of Station 1, and Station 4 is the downstream neighbor of Station 1.

Cabling Considerations for ATM LANE Modules

This section discusses topics that you should consider before connecting ATM LAN Emulation (LANE) modules to an ATM network.

Calculating Maximum Transmission Distances

When you use UTP Category 5 cabling for ATM LANE modules, note that the maximum transmission distance is 328 feet (100 meters). The maximum distances for ATM LANE fiber-optic network connections are determined by the transmitter output power, receiver sensitivity, and type of optical source. Table 2-10 lists the characteristics, including the maximum transmission distances, for both multimode and single-mode fiber-optic cables.

Connecting ATM Equipment

All ATM interfaces are full-duplex interfaces. You must use the appropriate ATM interface cable to connect the ATM multimode or UTP switching module with an external ATM network.

The ATM LANE switching modules provide an interface to ATM switching fabrics for transmitting and receiving data at rates of up to 155 Mbps (OC-3) bidirectionally; the actual rate is determined by the physical layer interface module (PLIM). The OC-12 dual PHY switching modules provide an interface to ATM switching fabrics for transmitting and receiving data at rates up to 622 Mbps (OC-12) bidirectionally.

The ATM LANE switching module can support PLIMs that connect to the following physical layers:

• Synchronous Optical Network (SONET) 155-Mbps MMF cable—Synchronous Transport Signal level 3, concatenated (STS-3c) or Synchronous Transfer Module level 1 (STM-1)
  
• SONET 155-Mbps SMF cable—STS-3c or STM-1
  
  
• UTP Category 5 RJ-45-STS-3C/STM-1
  

The ATM interface cable connects the switch to an ATM network or to two switches in series. You can obtain cables from these cable vendors:

• AT&T
  
  
• Siemens
  
  
• Red-Hawk
  
  
• Anixter
  
  
• AMP
  

For wide-area networking, ATM is currently being standardized in Broadband Integrated Services Digital Networks (BISDNs) by the International Telecommunication Union Telecommunication Standardization Sector (ITU-T) and ANSI. BISDN supports rates from E3 (34 Mbps) to multiple gigabits per second (Gbps).

The ATM User-Network Interface (UNI) specification defines the required Management Information Base (MIB) functionality for ATM interfaces. MIB attributes are readable and writable across the Integrated Local Management Interface (ILMI) using the Simple Network Management Protocol (SNMP). The ILMI uses SNMP together with the ATM ILMI MIB without User Datagram Protocol (UDP) or IP addressing to provide diagnostics, monitoring, and configuration services at the UNI.

The ATM LANE switching module supports RFC 1213 interface MIBs as specified in the ATM MIB V2 specification.

Cabling Considerations for an ATM Dual PHY DS3 Module

The ATM dual PHY DS3 module provides an interface to ATM switching fabrics for transmitting and receiving data at up to 45 Mbps bidirectionally.

This module uses 75-ohm RG-59 coaxial cable with bayonet-style twist-lock (BNC) connectors. Figure 2-15 shows the coaxial cable and BNC connectors.

The maximum station-to-station cabling distance for DS3 coaxial cable is 450 feet (137 meters).

Cabling Considerations for Token Ring Modules

When preparing your site for Token Ring connections, consider these factors:

• Type of cabling (fiber or twisted-pair cabling)
  
  
• Distance limitations
  
  
• Specific cables to connect each interface
  
  
• Any additional interface equipment, such as transceivers and converters
  
  

Setting Up Token Ring Cabling

The recommended distances for the various cable types are set by North American and international commercial building wiring standards. These standards state that standards-compliant horizontal copper cabling shall not exceed 295 feet (90 meters), leaving 33 feet (10 meters) total for required patch cabling in both the office and telecommunications closet.

Setting Up Cabling for Dedicated-Media LAN Segments

The Token Ring network dedicated-media connections support only one attached system per connection.

In a Token Ring network, a lobe is a section of cable that attaches a device to an access unit.

For all supported cable types except optical fiber, IEEE recommends a maximum cable lobe length of 625 feet (190 meters) plus a 33-feet (10 meter) total allowance for the patch cords in the office and the telecommunications closets. For optical fiber, IEEE recommends a maximum cable lobe length of 6562 feet (2000 meters).

Table 2-11 and Table 2-12 specify the maximum supported lobe lengths for the following cable types. Allow an additional 33 feet (10 meters) per lobe length to accommodate patch cables, unless otherwise specified.

• 150-ohm shielded media cable lobe lengths
  
  
• Lobe lengths for 100- or 120-ohm shielded or unshielded cable
  
  

.

Setting Up Cable Length and Lobe Wiring for Shared-Media LAN Segments

The cable types that you can use for shared-media LAN segments are the same as those described for dedicated-media segments. The hub or concentrator attached to the Catalyst 5000 family Token Ring module port determines the acceptable cable distances.

Determining Number of Attached Devices

A Token Ring network can support up to 260 attached devices or nodes on a single network when you use 150-ohm shielded media (type 1, 1A, 2, or 2A). When you use cable segments with 100- or 120-ohm shielded media, this number decreases to 132 (72 if you use any 4-Mbps-only adapters or filters).

Using Twisted-Pair Cable Pinouts

You must use a straight-through 100-ohm or 120-ohm cable when connecting devices to the Token Ring ports on the Catalyst 5000 family Token Ring module.

The RJ-45 connector on the Catalyst 5000 family Token Ring module makes ground available on the shield and on pins 1, 2, 7, and 8. Shielded cables provide continuity for the ground-to-any shielded connector on the other end of the cable.

Figure 2-16 and Figure 2-17 illustrate the straight-through 100-ohm and 120-ohm cable and the 150-ohm data connector-to-RJ-45 straight-through cable.

Connecting a Token Ring Network

To connect to a Token Ring network, use RJ-45 male connectors, as shown in Figure 2-18.

Connecting a Fiber Token Ring Module

The fiber Token Ring module provides 16 Volition VF-45 socket connections. Each Volition VF-45 connector supports an MMF connection to the Token Ring through a VF-45 to ST connector patch cable or VF-45 to SC connector patch cable. Figure 2-19 shows the Volition VF-45 connector.

Cabling Considerations for a Route Switch Module

For specific information on the RSM/VIP2 interface and cable requirements, refer to the Route Switch Module Catalyst VIP2-15 and VIP2-40 Installation and Configuration Note.