Issue No. 08 - August (1993 vol. 4)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.238625
<p>The authors describe parallel algorithms for simulating certain continuous time Markovchains, such as those arising in queueing network models of distributed computingsystems or communications systems. The algorithms are based on the technique ofuniformization. Two variations of a conservative parallel simulation algorithm arepresented. In each algorithm, a relatively short presimulation is performed to identifythose times, and only those times, at which the simulation algorithm requires processorpairs to synchronize. Speedup studies of the algorithms, performed on a 16-node InteliPSC/2 hypercube, are presented and discussed.</p>
Index Termscontinuous time Markov chains; uniformization; parallel algorithms; queueing networkmodels; distributed computing systems; communications systems; conservative parallelsimulation algorithm; Intel iPSC/2 hypercube; discrete event simulation; Markovprocesses; parallel algorithms; queueing theory
P. Heidelberger and D. Nicol, "Conservative Parallel Simulation of Continuous Time Markov Chains Using Uniformization," in IEEE Transactions on Parallel & Distributed Systems, vol. 4, no. , pp. 906-921, 1993.