About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
SIGMETRICS 1992
Conference paper
Analytical models of combining banyan networks
Abstract
We present in this paper an analytical model of a multistage combhting Banyan network with output buffered switches, in hot-spot traffic. In a combining network, packets bound for the same destination are combined into one if they meet at a switch; this alleviates the problem of tree-sat ur:ation caused by hot-spot traffic. We model the flow processes in the network as Markov chains and recursively approximatee the departure processes of each stage of the network in terms of the departure processes of the preceding stage. This model is used to predict the throughput of the combining network, and comparison with simulation results shows the prediction to be accurate. A modified combining scheme based on low priorities for hot packets is proposed and analyzed. It is shown that this scheme yields substantial improvements in throughput over the standard combining scheme.