This appendix provides a few tips on how you can increase the flexibility of your network if it is composed of multiple segments using mixed signaling methods. This appendix also outlines the basic properties of Class I and Class II repeaters and shows maximum cable lengths for various topologies.
The 100BASE-T technology defined in the IEEE 802.3u standard provides for both homogeneous Carrier Sense Multiple Access with Collision Detection (CSMA/CD) 100 Mbps networks and heterogeneous 10/100 Mbps mixed networks. Both network topologies can be supported by connecting various LAN segments using repeaters, switches, bridges, or routers.
The SuperStack II Hub 100 supports only 100 Mbps network topologies. Access to 10 Mbps functionality is possible through connections with switches, bridges, or routers that support 10/100 Mbps topologies. Full-duplex connections are not supported. Attached NICs and switches must be set to half-duplex.
Two LAN segments connected by a repeater constitute a single collision domain. LAN segments connected by switches, bridges, or routers constitute multiple collision domains. You can achieve maximum network flexibility by designing multiple collision domain networks.
For example, a combined system containing both 10BASE-T and 100BASE-T devices and built with repeaters and switches can deliver dedicated 100 Mbps, shared 100 Mbps, dedicated 10 Mbps, and shared 10 Mbps services to devices on the LAN. Figure B-1 illustrates a network composed of two collision domains connected by a 10/100 Mbps switching hub.
Figure B-1 Two Collision Domains Connected by a Switch
A Fast Ethernet packet transmitted on a LAN can pass through no more than two logical Class II repeaters or one logical Class I repeater before reaching its destination or passing through a LAN bridge, LAN switch, or router.
The IEEE 802.3u standard defines two classes of repeaters, Class I and Class II, for 100BASE-T Fast Ethernet networks, as detailed in the following sections. Table B-1 lists SuperStack hubs and shows their repeater classifications.
| 1
SuperStack II Hub 100 models described in this guide.
|
A Class I repeater is a hub with internal delay such that only one repeater can exist between any two end nodes within a single collision domain when two maximum-length cable segments are used.
A Class I network topology consists of one hub (or hub stack) in a single collision domain existing between any two end nodes. A stack of hubs in this configuration can contain up to eight SuperStack II Hub 100 T4 units, eight SuperStack II Hub 100 TX units (Class II repeaters can be used in a Class I network topology), or a combination of Hub 100 T4 and Hub 100 TX units.
Table B-2 provides maximum cable lengths between a Class I repeater and an end node in a single collision domain for various cable types and connectors.
Figure B-2 shows a Class I repeater configuration using only UTP cable. In this configuration, the stack connects end nodes in a maximum network diameter of 200 meters. The end nodes and the hub stack are in the same collision domain.
The figures in this appendix use the following terms to describe cable types: TX UTP, T4 UTP, and FX fiber. These abbreviations refer to cables that conform to the 100BASE-TX, 100BASE-T4, and 100BASE-FX IEEE 802.3 specifications for 100 Mbps Ethernet signaling. See "100BASE-T Standard" in Appendix A for details.
Figure B-2 Class I Hub Using TX and T4 UTP Cable
If a 100BASE-FX transceiver interface module is used in combination with T4 UTP cable, the total span between end nodes can be 231 meters. That is, there can be up to 100 meters of T4 UTP cable and up to 131 meters of fiber-optic cable, as shown in Figure B-3.
Figure B-3 Class I Hub Using FX Fiber and T4 UTP Cable
A Class II repeater is a hub with internal delay such that no more than two hubs can exist between any two end nodes within a single collision domain when two maximum-length cable segments are used. The Class II network topology allows two hubs (or hub stacks) to exist between any two end nodes.
The following SuperStack II Hub 100 TX models can operate as Class II repeaters: 3C250C-TX-12, 3C250C-TX-24, and 3C250B-TX. These models replace the SuperStack II Hub 100 TX (part number 3C250A-TX / I), which can operate only as a Class I repeater.
Table B-3 provides maximum cable lengths between a Class II repeater and an end node or between two Class II repeaters in a single collision domain for various cable types and connectors.
| 1
Assumes 105 meters of copper link and one fiber-optic link.
|
Figure B-4 shows a single-repeater collision domain containing one Class II hub between end nodes that are located 200 meters apart. This length is the total allowable span for UTP cable.
Figure B-4 Class II Hub Using TX UTP Cable
If a 100BASE-FX transceiver interface module is used in combination with TX UTP cable, the total span between end nodes can be 308.8 meters. That is, there can be up to 100 meters of TX UTP cable and up to 208.8 meters of fiber-optic cable (see Figure B-5).
Figure B-5 Class II Hub Using FX Fiber and TX UTP Cable
Figure B-6 shows two Class II repeaters sharing a collision domain. The repeaters are placed between end nodes that are located 205 meters apart, which is the total allowable span for UTP cable.
Figure B-6 Class II Hubs Using TX UTP Cable
If a 100BASE-FX transceiver interface module is used in combination with TX UTP cable, the total span between end nodes for two Class II hubs sharing one collision domain can be 216.2 meters. That is, there can be up to 105 meters of TX UTP cable and up to 111.2 meters of fiber-optic cable (see Figure B-7).
Figure B-7 Class II Hubs Using FX Fiber and TX UTP Cable
![[previous]](../../../graphics/prev.gif)
![[next]](../../../graphics/next.gif)