Optimizing cache access: A tool for source-to-source transformations and real-life compiler tests

Research output: Contribution to book/Conference proceedings/Anthology/ReportChapter in book/Anthology/ReportContributedpeer-review

Contributors

  • Ralph Müller-Pfefferkorn - , Center of Information Services and High Performance Computing (ZIH), TUD Dresden University of Technology (Author)
  • Wolfgang E. Nagel - , TUD Dresden University of Technology (Author)
  • Bernd Trenkler - , TUD Dresden University of Technology (Author)

Abstract

Loop transformations are well known to be a very useful tool for performance improvements by optimizing cache access. Nevertheless, the automatic application is a complex and challenging task especially for parallel codes. Since the end of the 1980's it has been promised by most compiler vendors that these features will be implemented - in the next release. We tested current FORTRAN90 compilers (on IBM, Intel and SGI hardware) for their capabilities in this field. This paper shows the results of our analysis. Motivated by this experience we have developed the optimization environment Goofi to assist programmers in applying loop transformations to their code thus gaining better performance for parallel codes even today.

Details

Original languageEnglish
Title of host publicationEuro-Par 2004 Parallel Processing
EditorsMarco Danelutto, Marco Vanneschi, Domenico Laforenza
PublisherSpringer, Berlin [u. a.]
Pages72-81
Number of pages10
ISBN (print)3540229248
Publication statusPublished - 2004
Peer-reviewedYes
Externally publishedYes

Publication series

SeriesLecture Notes in Computer Science, Volume 3149
ISSN0302-9743

Keywords