An optimal service scheduling policy for packet networks with quality of service guarantees

In this paper, we propose a measurement-based service scheduling policy, called the D-policy, for high-speed packet networks with quality of service guarantees. We show that the D-policy is optimal with respect to channel efficiency among all stable scheduling policies. We examine the properties of the D-policy and show how these properties can be used to compute the minimum bandwidth required for a given traffic scenario. We also compare the D-policy with other policies such as the first-come-first-served, static priority, and weighted round robin through simulation studies.