From serial loops to parallel execution on distributed systems

George Bosilca, Aurelien Bouteiller, Anthony Danalis, Thomas Herault, Jack Dongarra

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

3 Scopus citations

Abstract

Programmability and performance portability are two major challenges in today's dynamic environment. Algorithm designers targeting efficient algorithms should focus on designing high-level algorithms exhibiting maximum parallelism, while relying on compilers and run-time systems to discover and exploit this parallelism, delivering sustainable performance on a variety of hardware. The compiler tool presented in this paper can analyze the data flow of serial codes with imperfectly nested, affine loop-nests and if statements, commonly found in scientific applications. This tool operates as the front-end compiler for the DAGuE run-time system by automatically converting serial codes into the symbolic representation of their data flow. We show how the compiler analyzes the data flow, and demonstrate that scientifically important, dense linear algebra operations can benefit from this analysis, and deliver high performance on large scale platforms.

Original languageEnglish
Title of host publicationParallel Processing - 18th International Conference, Euro-Par 2012, Proceedings
Pages246-257
Number of pages12
DOIs
StatePublished - 2012
Externally publishedYes
Event18th International Conference on Parallel Processing, Euro-Par 2012 - Rhodes Island, Greece
Duration: Aug 27 2012Aug 31 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7484 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Parallel Processing, Euro-Par 2012
Country/TerritoryGreece
CityRhodes Island
Period08/27/1208/31/12

Keywords

  • compiler analysis
  • distributed computing
  • symbolic data flow
  • task scheduling

Fingerprint

Dive into the research topics of 'From serial loops to parallel execution on distributed systems'. Together they form a unique fingerprint.

Cite this