GPUexplore 3.0: GPU Accelerated State Space Exploration for Concurrent Systems with Data

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

21 Downloads (Pure)

Abstract

GPUexplore 3.0 is an explicit state space exploration tool that runs entirely on a graphics processing unit (GPU), and supports models of concurrent systems with data variables. We discuss its workflow and modelling language, present several design decisions regarding work distribution and retrieval, and experimentally evaluate the impact of those decisions. Our tool achieves acceleration up to 115 × and 28 × compared to single- and four-core LTSmin, respectively. It currently checks for deadlocks, with verification of temporal logic formulae planned for the near future.

Original languageEnglish
Title of host publicationModel Checking Software
Subtitle of host publication29th International Symposium, SPIN 2023, Paris, France, April 26–27, 2023, Proceedings
EditorsGeorgiana Caltais, Christian Schilling
Place of PublicationCham
PublisherSpringer
Pages188-197
Number of pages10
ISBN (Electronic)978-3-031-32157-3
ISBN (Print)978-3-031-32156-6
DOIs
Publication statusPublished - 2 May 2023
Event29th International Symposium on Model Checking Software, SPIN 2023 - Paris, France
Duration: 26 Apr 202327 Apr 2023

Publication series

NameLecture Notes in Computer Science (LNCS)
Volume13872
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference29th International Symposium on Model Checking Software, SPIN 2023
Abbreviated titleSPIN 2023
Country/TerritoryFrance
CityParis
Period26/04/2327/04/23

Keywords

  • Explicit state space exploration
  • GPU
  • finite-state machines

Fingerprint

Dive into the research topics of 'GPUexplore 3.0: GPU Accelerated State Space Exploration for Concurrent Systems with Data'. Together they form a unique fingerprint.

Cite this