Distributed stochastic discrete-event simulation in parallel time streams

Abstract. Quantitative stochastic simulation as a tool used by engineers of various disciplines for studying and evaluating performance of various systems suffers from the fact that sound simulation studies requires very long runlength to obtain the results with the accuracy. In this paper we look at traditional approaches to distributed quantitative stochastic simulation and propose a new scenario, named Multiple Replications in Paralel Time Streams (MRIP), that solve the problem in an efficient way. An implementation of MRIP in a simulation package AKAROA has been also described. AKAROA accepts ordinary (nonparallel) simulation models and creates fully automatically the environment required for running MRIP on workstations of a local area network. Presented results show that MRIP offers linear speedup of simulation. Limitations of this scenario for running distributed quantitative stochastic simulation are also discussed. 1.