Answers are grounded in this paper and cite their sources.
In this paper we analyze the performance of a preemptive priority queue. We give the model description in the context of a packet communication system where message sources, having different priorities, share a common communication channel. Each source generates, as an independent Poisson process, messages consisting of an arbitrarily distributed, random number of fixed-length packets. The channel server can only begin service at integer multiples of the packet transmission time (i.e., a time-slotted channel is assumed), and the server will preempt an ongoing message transmission at the next packet boundary whenever there is a message arrival from a higher-priority source. The average in-queue waiting time for each packet in any given source message and the average message delay are derived along with the corresponding moment-generating functions. Also, comparisons are made with the first-come first-served queueing discipline.
No takes yet. Share an insight, caveat, or question.
Hluchyj et al. (1983) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: