Automatic Discovery of Parallelism: A Tool and an Experiment (Extended Abstract)

  • Michael Burke
  • , Ron Cytron
  • , Jeanne Ferrante
  • , Wilson Hsieh
  • , Vivek Sarkar
  • , David Shields

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

This paper reports preliminary results from applying advanced techniques to the parallelization of sequential programs. Such techniques include interprocedural analysis and the identification of nested parallelism. These techniques have been proposed for exploiting the greater concurrency offered by multiprocessors as compared with vector architectures. The effectiveness of these techniques as applied to some popular numerical Fortran programs is examined. Although extrapolation is difficult, these preliminary results are encouraging indicators of the success of these techniques.

Original languageEnglish
Pages (from-to)77-84
Number of pages8
JournalSIGPLAN Notices (ACM Special Interest Group on Programming Languages)
Volume23
Issue number9
DOIs
StatePublished - Jan 1 1988

Fingerprint

Dive into the research topics of 'Automatic Discovery of Parallelism: A Tool and an Experiment (Extended Abstract)'. Together they form a unique fingerprint.

Cite this