Optical Networks: A Practical Perspective - Part 58. This book describes a revolution within a revolution, the opening up of the capacity of the now-familiar optical fiber to carry more messages, handle a wider variety of transmission types, and provide improved reliabilities and ease of use. In many places where fiber has been installed simply as a better form of copper, even the gigabit capacities that result have not proved adequate to keep up with the demand. The inborn human voracity for more and more bandwidth, plus the growing realization that there are other flexibilities to be had by imaginative use of the fiber, have led people. | 540 Network Survivability SONET SDH. A ring is the simplest topology offering an alternate route around a failure. In the optical layer many protection schemes have been designed to operate over true mesh topologies. Protection may be dedicated or shared. In dedicated protection each working connection is assigned its own dedicated bandwidth in the network over which it can be rerouted in case of a failure. In shared protection we make use of the fact that not all working connections in the network fail simultaneously for example if they are in different parts of the network . Therefore by careful design we can make multiple working connections share protection bandwidth among themselves. This helps reduce the amount of bandwidth needed in the network for protection. Another advantage of shared protection is that the protection bandwidth is available to carry low-priority traffic under normal conditions. This low-priority traffic is discarded in the event of a failure when the bandwidth is needed to protect a connection. Protection schemes can either be revertive or nonrevertive. In both schemes if a failure occurs traffic is switched from the working path to the protect path. In a nonrevertive scheme the traffic remains on the protect path until it is manually switched back onto the original working path usually by a user through the network management system. In a revertive scheme once the working path is repaired the traffic is automatically switched back from the protect path onto the working path. Reversion allows the network to return to its original state once the failure is restored. Dedicated protection schemes may be revertive or nonrevertive however shared protection schemes are usually revertive. Since multiple working connections share a common protection bandwidth the protection bandwidth must be freed up as soon as possible after the original failure has been repaired so that it can be used to protect other connections in the event of another failure
