Skip to main navigation Skip to search Skip to main content

An optimization method for embarrassingly parallel under MIC architecture

  • Beihang University

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

Abstract

Nowadays, heterogeneous architecture of CPU plus accelerator has become a mainstream in supercomputing. Intel lauched its Xeon Phi coprocessor in this context. It uses Intel's many-core architecture, which greatly improves the single node parallelism. This paper studies the optimization of embarrassingly parallel programs under Intel MIC architecture, to maximize the utilization of CPU and Phi processor, and reduce the running time of parallel programs, by combining the computing power of CPU and Phi. This so-called embarrassingly parallel program often have do all main loops, that is, there are no dependencies between iterations, so they can be fully parallelized. This do all loop exists in many typical parallel programs. We come up with a loop allocation method for do all loops under this CPU/MIC architecture, to satisfy the above performance objectives.

Original languageEnglish
Title of host publicationProceedings - 14th International Symposium on Distributed Computing and Applications for Business, Engineering and Science, DCABES 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages17-20
Number of pages4
ISBN (Electronic)9781467365932
DOIs
StatePublished - 8 Mar 2016
Event14th International Symposium on Distributed Computing and Applications for Business, Engineering and Science, DCABES 2015 - Guiyang, China
Duration: 18 Aug 201524 Aug 2015

Publication series

NameProceedings - 14th International Symposium on Distributed Computing and Applications for Business, Engineering and Science, DCABES 2015

Conference

Conference14th International Symposium on Distributed Computing and Applications for Business, Engineering and Science, DCABES 2015
Country/TerritoryChina
CityGuiyang
Period18/08/1524/08/15

Keywords

  • Embarrassingly parallel
  • Exascale
  • Intel Xeon Phi
  • Loop allocation
  • Many-core
  • Performance tuning

Fingerprint

Dive into the research topics of 'An optimization method for embarrassingly parallel under MIC architecture'. Together they form a unique fingerprint.

Cite this