EVALUATING SCHEDULING IN INTEGRATED-SERVICES NETWORKS USING A SEMI-FLUID-FLOW MODEL

An integrated-services network carries diverse traffic, which leads to diverse performance objectives. For example, voice and video packets typically have performance objectives based on the fraction of packets that will be delivered within a given delay bound, while data packets often have objectives based on mean delay. Greater loads can be suppo rted in networks in which a voice or video packet is given priority ov er data packets if and only if the former is in danger of missing its deadline. Algorithms that allow this include Cost-Based Scheduling, oc cupancy-based algorithms, the Priority Token Bank, and to a lesser ext ent, the Leaky Bucket. This paper presents an approach to evaluating p erformance with these algorithms in the realistic case where data arri vals are highly bursty, but voice and video packets are not. Mean queu eing delay for data bursts is determined analytically in some importan t scenarios, and an efficient simulation approach based on the same mo del is described for cases where analysis is not currently possible. T he model is a semi-fluid-flow model in which voice or video packets ar e assumed to arrive as a continuous fluid dow, whereas data packets ar rive in large bursts at discrete instants in time.