The Utilization of Priorities on Token Ring NetworksReport
Many local area networks supply the user with multiple packet priority levels for network access. The intent is to discriminate in favor of those messages which the sender deems most important. This discrimination between higher and lower priority messages is realized as a difference in their end-to-end delivery delays, with higher priority messages having less delay than lower priority messages. There are, however, problems associated with priority schemes. Among these are the overhead generated by prioritizing messages, and the determination of which messages belong in each priority level. If the correct decisions are not made, optimal network performance will not result. Our thesis is that, in the nonnal environment where the network is not overloaded, no packet priority levels should be implemented (i.e., exactly one message class should be used); furthermore, using multiple packet priorities in this environment will generally degrade overall performance. For priority operation to be meaningful, priority service must be given at the point of the service bottleneck. Our experience is that the bottleneck is not network access, but rather the software processing above network access (e.g., OSI protocols). If a message has not been treated expeditiously prior to its bid for network access, assigning a high network access priority does little to decrease its latency.
Note: Abstract extracted from PDF file via OCR
All rights reserved (no additional license for public reuse)
Peden, Jeffery, and Alfred Weaver. "The Utilization of Priorities on Token Ring Networks." University of Virginia Dept. of Computer Science Tech Report (1988).
University of Virginia, Department of Computer Science