Programming heterogeneous parallel machines using refactoring and Monte–Carlo tree search.

Brown, Christopher; Janjic, Vladimir; Goli, M.; McCall, J.

Abstract

This paper presents a new technique for introducing and tuning parallelism for heterogeneous shared-memory systems (comprising a mixture of CPUs and GPUs), using a combination of algorithmic skeletons (such as farms and pipelines), Monte–Carlo tree search for deriving mappings of tasks to available hardware resources, and refactoring tool support for applying the patterns and mappings in an easy and effective way. Using our approach, we demonstrate easily obtainable, significant and scalable speedups on a number of case studies showing speedups of up to 41 over the sequential code on a 24-core machine with one GPU. We also demonstrate that the speedups obtained by mappings derived by the MCTS algorithm are within 5–15% of the best-obtained manual parallelisation.

Citation

BROWN, C. JANJIC, V., GOLI, M. and MCCALL, J. 2020. Programming heterogeneous parallel machines using refactoring and Monte–Carlo tree search. International journal of parallel programming [online], 48(4): high level parallel programming, pages 583-602. Available from: https://doi.org/10.1007/s10766-020-00665-z

• Journal Article Type: Article
• Acceptance Date: May 27, 2020
• Online Publication Date: Jun 10, 2020
• Publication Date: Aug 31, 2020
• Deposit Date: Jun 26, 2020
• Publicly Available Date: Jun 26, 2020
• Journal: International journal of parallel programming
• Print ISSN: 0885-7458
• Electronic ISSN: 1573-7640
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Volume: 48
• Issue: 4
• Pages: 583-602
• DOI: https://doi.org/10.1007/s10766-020-00665-z
• Keywords: Heterogeneous parallel computing; Monte–Carlo tree search; Optimisations
• Public URL: https://rgu-repository.worktribe.com/output/937862