Hubs

In the beginning of Ethernet, 10Base-5 used a very thick cable that was hard to work with (it was nicknamed thicknet). 10Base-2, which latter replaced 10Base-5, used a much smaller cable, similar to that used for cable TV. 10Base-2 was nickname thin-net. These cable technologies required large metal couplers called N connectors (10Base-5) and BNC connectors (10Base-2). These Networks also required special terminators to be installed at the end of cable runs. When these couplers or terminators were removed. the entire network would stop working. These cables formed the physical backbones for Ethernet networks.

With the introduction of Ethernet running over unshielded twisted pair (UTP) cables terminated with RJ45 connectors, hubs became the new backbones in most installations. Many companies attached hubs to their existing thin-net networks to allow greater flexibility as well. Hubs were made to support UTP and BNC 10Base-2 installations but UTP was so much easier to work with that it became the de factor standard.

A Hub is simple a means of connecting Ethernet cables together so that their  signals can be repeated to every other connected cable on the hub. Hubs may also be called repeaters for this reason, but it is important to understand that while a hub is a repeater, a repeater is not necessarily a hub.

A repeater repeats a signal. Repeaters are usually used to extend a connection to a remote host, or to connect a group of users who exceed the distance limitation of 10Base-T. In other words, if a usable distance of a 10Base-T cable is exceeded, a repeater can be placed in line to increase the usable distance.

Segments are divided by repeaters or hubs. Figure 2-1 shows a repeater extending the distance between a server and a personal computer.

A hub is like a repeater, except that while a repeater may have only two connectors, a hub can have many more; that is, it repeats a signal over many cables as opposed to just one. Figure 2-2 shows a hub connecting several computers to a network.

When designing Ethernet networks, repeater and hubs get treated the same way. The 5-4-3 Rule of Ethernet design states that between any two nodes on an Ethernet network, there can be only five segments, connected via four repeaters, and only three of the segments can be populated. This rule, which seems odd in the context of today's networks, was the source of much pain for those who didn't understand it.

A hubs became less expensive, extra hubs were often used as repeaters in more complex networks. Figure 2-3 shows an example of how two remote groups of users could be connected using hubs on each end and a repeater in the middle.

Hubs are very simple devices. Any signal received on any port is repeated out every other port. Hubs are purely physical and electrical devices, and do not have a presence on the network (Except possibly for management purposes). They do not alter frames or make decisions based on them in any way.

Figure 2-4 illustrates how hubs operate. As you might imagine, this model can become problematic in larger networks. The traffic can become so intensive that the network becomes saturated--if someone prints a large file, everyone on the network will suffer while the file is transferred to the printer over the network.

If another device is already using the wire, the sending device will wait a bit, and then try to transmit again. When two stations transmit at the same time, a collision occurs. Each station records the collision, backs off again, and then retransmits. On every busy networks, a lot of collisions will occur.

With a hub, more stations are capable of using the network at any given time. Should all of the stations be active, the network will appear to be slow because of the excessive collisions.

Collisions are limited to network segments. An Ethernet network segment is a section of network where devices can communicate using layer-2MAC addresses. To communicate outside of an Ethernet segment, an additional device, such as a router, is required. Collisions are also limited to collision domains. A collision domain is an area of an Ethernet network where collisions can occur. If one station can prevent another from sending because it has the network in use, these stations are in the same collision domain.

A broadcast domain is the area of an Ethernet network where a broadcast will be propagated. Broadcasts stay within a layer-3 network (unless forwarded), which is usually bordered by a layer-3 device such as a router. Broadcasts are sent through switches (layer-2 devices), but stop at router.

Figure 2-5 shows a network of hubs connected via a central hub. when a frame enters the hub on the bottom left on port 1, the frame is repeated out every other port on that hub, which includes a connection to the central hub. The central hub is turn repeats the frame out every port, propagating it to the remaining hubs in the network. This design replicates the backbone idea, in that every device on the network will receive every frame sent to the network.

   

In large networks of this type, new problems can arise. Late Collisions occur when two stations successfully test for a clear network, and then transmit, only to then encounter a collision. This condition can occur when the network is so large that the propagation of a transmitted frame from one end of the network to the other takes longer than the test used to detect whether the network is clear.

One of the major problems when using hubs is the possibility of broadcast storms. Figure 2-6 shows the situation.

The only way to resolve a broadcast storm is to break the loop. Shutting down and restarting the network devices will just start the cycle again. Because hubs are not generally manageable, it can be quite a challenge to find a layer-2 loop in a crisis.

Hubs have a lot of drawbacks, and modern networks rarely employ them. Hubs have long since been replaced by switches, which offer greater speed, automatic loop detection, and a host of additional features.

Source:: O'Reilly-Network Warrior