Token Ring is an old networking topology that worked very well but unfortunately was a lot more expensive due to the licensing issues (mainly because the hardware required licensing from IBM)
Essentially, in the early days it was a physical “ring” of wire that was connected to a central point. Each system took a turn talking on the network when the <i>token </i>came around to it. Once it completed communicating it would pass the token to the next system in the ring. This continued for as long as the ring was kept constant— unfortunately, if a user unplugged their cable from their system it would break the ring and teh whole ring would go down.
Typically, because of this issue administrators tried to limit the size of a ring by bringing multiple rings back to a single location. Still farther down the historical line switches were produced that ran over twisted pair cables with RJ-45 connections that made the ring the size of one PC.
Another common problem with token ring was sometimes when a system failed it would “<i>beacon</i>” on the ring. This was to let the administrator know what system was failing– however, a beaconing system would bring the ring down.
As a general rule, Token ring was good for small packet consistant sizes that were transmitted on a regular basis, but was not so good for large inconsistent packet sizes transmitted over a period of time sporatically– like video and large files.
Here are some links to some background on it: