Efficient dynamic pinning of parallelized applications by reinforcement learning with applications

Georgios C. Chasparis (Lead / Corresponding author), Michael Rossbory, Vladimir Janjic

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

Abstract

This paper describes a dynamic framework for mapping the threads of parallel applications to the computation cores of parallel systems. We propose a feedback-based mechanism where the performance of each thread is collected and used to drive the reinforcement-learning policy of assigning affinities of threads to CPU cores. The proposed framework is flexible enough to address different optimization criteria, such as maximum processing speed and minimum speed variance among threads. We evaluate the framework on the Ant Colony optimization parallel benchmark from the heuristic optimization application domain, and demonstrate that we can achieve an improvement of 12% in the execution time compared to the default operating system scheduling/mapping of threads under varying availability of resources (e.g. when multiple applications are running on the same system).

Original languageEnglish
Title of host publicationEuro-Par 2017
Subtitle of host publicationParallel Processing - 23rd International Conference on Parallel and Distributed Computing, Proceedings
EditorsFrancisco F. Rivera, Tomas F. Pena, Jose C. Cabaleiro
Place of PublicationSwitzerland
PublisherSpringer
Pages164-176
Number of pages13
ISBN (Electronic)9783319642031
ISBN (Print)9783319642024
DOIs
Publication statusPublished - 2017
Event23rd International Conference on Parallel and Distributed Computing, Euro-Par 2017 - Santiago de Compostela, Spain
Duration: 28 Aug 20171 Sep 2017

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume10417
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference23rd International Conference on Parallel and Distributed Computing, Euro-Par 2017
CountrySpain
CitySantiago de Compostela
Period28/08/171/09/17

Fingerprint Dive into the research topics of 'Efficient dynamic pinning of parallelized applications by reinforcement learning with applications'. Together they form a unique fingerprint.

  • Profiles

    No photo of Vladimir Janjic

    Janjic, Vladimir

    Person: Academic

    Cite this

    Chasparis, G. C., Rossbory, M., & Janjic, V. (2017). Efficient dynamic pinning of parallelized applications by reinforcement learning with applications. In F. F. Rivera, T. F. Pena, & J. C. Cabaleiro (Eds.), Euro-Par 2017: Parallel Processing - 23rd International Conference on Parallel and Distributed Computing, Proceedings (pp. 164-176). (Lecture Notes in Computer Science; Vol. 10417). Springer . https://doi.org/10.1007/978-3-319-64203-1_12