An Experimental Evaluation of Probabilistic Simulation

  • Authors:
  • Jonathan Bogdoll;Holger Hermanns;Lijun Zhang

  • Affiliations:
  • Department of Computer Science, Saarland University, Saarbrücken, Germany;Department of Computer Science, Saarland University, Saarbrücken, Germany;Department of Computer Science, Saarland University, Saarbrücken, Germany

  • Venue:
  • FORTE '08 Proceedings of the 28th IFIP WG 6.1 international conference on Formal Techniques for Networked and Distributed Systems
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Probabilisticmodel checking has emerged as a versatile system verification approach, but is frequently facing state-space explosion problems. One promising attack to this is to construct an abstract model which simulatesthe original model, and to perform model checking on that abstract model. Recently, efficient algorithms for deciding simulation of probabilistic models have been proposed. They reduce the theoretical complexity bounds drastically by exploiting parametric maximum flow algorithms. In this paper, we report on experimental comparisons of these algorithms, together with various interesting optimizations. The evaluation is carried out on both standard PRISM example cases as well as randomly generated models. The results show interesting time-space tradeoffs, with the parametric maximum flow algorithms being superior for large, dense models.