Year: 1991
Type: conference-paper
Abstract: In simulating applications for execution on specific computing systems, the simulation performance figures must be known in a short period of time. One basic approach to the problem of reducing the required simulation time is the exploitation of parallelism. However, in parallelizing the simulation new problems arise. Due to the distributed generation of events causality errors can occur, as a result the sequence in which to process the events is essentially indeterminated. In this paper we present a model to analyse the inherent parallelism of a simulation, together with a survey of existing strategies to perform the simulation in parallel. Some extensions to this model are discussed, resulting in reliable evaluation of the effectiveness of these strategies. 1 Introduction In the Parallel Scientific Computing Working-group at the University of Amsterdam, we are interested in the execution performance of classes of applications on classes of computing systems. We distinguish the follo...
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2015315991