In a ring network, every device has exactly two neighbours for communication purposes. All messages travel through a ring in the same direction. Like the bus topology, a failure in any cable or device breaks the loop and will take down the entire segment.
A disadvantage of the ring is that if any device is added to or removed from the ring, the ring is broken and the segment fails until it is reconnected. It is also considerably more expensive than other topologies.
Ring topology is relatively rare.
Ring topologies use Token Ring technology to manage permissions for nodes to transmit.
The token constantly circles the ring until a node needs to transmit. When the token arrives at the node, the node captures the token and attaches its message because holding the empty token gives the node permission to speak. It attaches a destination address to the token and sends it on its way.
As the token and message arrive at each node, the nodes examines the token to see if it is addressed to them. If a node finds the token is not addressed to them, the node passes the token along. Even if the node wished to transmit, it would not be allowed to even though it held the token, because the token had a message attached to it.
When the token and message arrive at the destination, the node sees the message is addressed to them and reads it. The node then passes the token and message onward to the next node.
When the token and message arrive at the node who sent originally sent it, it removes the message and passes the empty token downstream. The next node that wished to speak could then grab the token and attach its own message.
In this way, every node is guaranteed a chance to speak - eventually and there is no chance of more than one node trying to speak at the same time causing a network "collision."
|