GPGPU applied to support the construction of the state-space graphs of IOPT Petri net models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Graphics Processing Units (GPU) for General-Purpose parallel computing are applied in this work to support the construction of state-space graphs of Input-Output Place-Transition (IOPT) Petri net models (commonly known as reachability graphs). Starting from previous works already integrated and publicly available in the IOPT-Tools framework, a new algorithm to build the state-space graph is proposed based on its adaptation to General-Purpose computing on GPU platforms (GPGPU). To implement the new algorithm, part of the code that is automatically generated by the state-space generator of the IOPT tool framework was adapted to run on a NVIDIA GPU under the Compute Unified Device Architecture (CUDA) Toolkit. The GPU was used as a co-processor, namely the known sequential part of the algorithm runs on the CPU and the identified computationally-intensive part, concerning to the treatment of each unprocessed state and the calculation of all its child nodes, is handled by the GPU. A set of IOPT models already available at the IOPT-Tools framework are used to validate the results obtained with the new algorithm.

Original languageEnglish
Title of host publicationProceedings: IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society
PublisherIEEE Computer Society
Pages5862-5867
Number of pages6
ISBN (Electronic)9781728148786
DOIs
Publication statusPublished - Oct 2019
Event45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019 - Lisbon, Portugal
Duration: 14 Oct 201917 Oct 2019

Publication series

NameIECON Proceedings (Industrial Electronics Conference)
PublisherIEEE Computer Society
Volume2019-October

Conference

Conference45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019
CountryPortugal
CityLisbon
Period14/10/1917/10/19

Fingerprint Dive into the research topics of 'GPGPU applied to support the construction of the state-space graphs of IOPT Petri net models'. Together they form a unique fingerprint.

Cite this