The Stochastic Rendezvous Network Model for Performance of Synchronous Client-Server-like Distributed Software
Issue No. 01 - January (1995 vol. 44)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/12.368012
<p><it>Abstract—</it>Distributed or parallel software with synchronous communication via rendezvous is found in client-server systems and in proposed Open Distributed Systems, in implementation environments such as Ada, V, Remote Procedure Call systems, in Transputer systems, and in specification techniques such as CSP, CCS and LOTOS. The delays induced by rendezvous can cause serious performance problems, which are not easy to estimate using conventional models which focus on hardware contention, or on a restricted view of the parallelism which ignores implementation constraints. Stochastic Rendezvous Networks are queueing networks of a new type which have been proposed as a modelling framework for these systems. They incorporate the two key phenomena of <it>included service</it> and a <it>second phase of service</it>. This paper extends the model to also incorporate different services or <it>entries</it> associated with each task. Approximations to arrival-instant probabilities are employed with a Mean-Value Analysis framework, to give approximate performance estimates. The method has been applied to moderately large industrial software systems.</p><p><it>Index Terms—</it>Client-server, performance, remote procedure call, software performance, distributed software, rendezvous networks, multiple entries.</p>
D. C. Petriu, J. E. Neilson, a. S. Majumdar and C. M. Woodside, "The Stochastic Rendezvous Network Model for Performance of Synchronous Client-Server-like Distributed Software," in IEEE Transactions on Computers, vol. 44, no. , pp. 20-34, 1995.