Skip to content

Enabling region merging optimizations in OpenMP

1Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Maximizing the scope of a parallel region, which avoids the costs of barriers and of launching additional parallel regions, is among the first recommendations in any optimization guide for OpenMP. While clearly beneficial and easily accomplished for code where regions are visibly contiguous, regions often become contiguous only after compiler optimization or resolution of abstraction layers. This paper explores changes to the OpenMP specification that would allow implementations to merge adjacent parallel regions automatically, including the removal of issues that make the transformation non-conforming and the addition of hints that facilitate the optimization. Beyond simple merging, we explore hints to fuse workshared loops that occur in syntactically distinct parallel regions or to apply nowait to such loops. Our evaluation shows these changes can provide an overall speedup of 2–8× for a microbenchmark, or 6 % for a representative physics application.

Cite

CITATION STYLE

APA

Scogland, T. R. W., Gyllenhaal, J., Keasler, J., Hornung, R., & de Supinski, B. R. (2015). Enabling region merging optimizations in OpenMP. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9342, pp. 177–188). Springer Verlag. https://doi.org/10.1007/978-3-319-24595-9_13

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free