GPU Accelerating Statistical Model Checking for Extended Timed Automata.

Oliver S. Bak, Mathias W. B. Christiansen, Oliver V. Eriksen, Sergio Feo-Arenis, Peter Gjøl Jensen, Marcus D. Jensen, Simas Juozapaitis, Kim G. Larsen, Marius Mikucionis, Marco Antonio Muniz Rodriguez, Andreas Windfeld

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

The core component of Statistical Model Checking (SMC) is the repeated sampling of a given system as to estimate statistical measures. To obtain probabilistic estimates with high confidence a significant number of simulations is required, in particular in the presence of rare events. In this paper we explore the use of Graphical Processing Unit (GPU) for accelerating SMC for Networks of Stochastic Extended Timed Automata (SXTA). We discuss the many challenges and solutions required to achieve significant speedups on a GPU architecture. In collaboration with NVIDIA we develop a prototype tool for parallel SMC using both GPU and multi-core CPU. Experimental results demonstrate trade-offs in the computation time when utilizing either CPU or GPU. In one case we observed the GPU using 20% of the power of the CPU equivalent while delivering a 2.73 time speedup.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number of pages26
Publication date2025
Pages267-292
DOIs
Publication statusPublished - 2025

Bibliographical note

DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.

Fingerprint

Dive into the research topics of 'GPU Accelerating Statistical Model Checking for Extended Timed Automata.'. Together they form a unique fingerprint.

Cite this