Time-Efficient model checking with magnetic disk

  • Authors:
  • Tonglaga Bao;Michael Jones

  • Affiliations:
  • Dept. of Computer Science, Brigham Young U., Provo, Utah;Dept. of Computer Science, Brigham Young U., Provo, Utah

  • Venue:
  • TACAS'05 Proceedings of the 11th international conference on Tools and Algorithms for the Construction and Analysis of Systems
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Explicit model checking with magnetic disk is prohibitively slow if file input/output (IO) is not carefully managed. We give an empirical analysis of the two published algorithms for model checking with magnetic disk and show that both algorithms minimize file IO time but are dominated by delayed duplicate detection time (which is required to avoid regenerating parts of the transition graph). We present and analyze a more time-efficient algorithm for model checking with magnetic disk that requires more file IO time, but less delayed duplicate detection time and less total execution time. The new algorithm is a variant of parallel partitioned hash table algorithms and uses a time-efficient chained hash table implementation.