Advanced search
1 file | 1.04 MB

Factoring out ordered sections to expose thread-level parallelism

Hans Vandierendonck (UGent) , Sean Rul (UGent) and Koen De Bosschere (UGent)
Author
Organization
Abstract
With the rise of multi-core processors, researchers are taking a new look at extending the applicability auto-parallelization techniques. In this paper, we identify a dependence pattern on which autoparallelization currently fails. This dependence pattern occurs for ordered sections, i.e. code fragments in a loop that must be executed atomically and in original program order. We discuss why these ordered sections prohibit current auto-parallelizers from working and we present a technique to deal with them. We experimentally demonstrate the efficacy of the technique, yielding significant overall program speedups.
Keywords
Multi-core, ordered sections, auto-parallelization

Downloads

  • Vandierendonck-Paper11.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 1.04 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Vandierendonck, Hans, Sean Rul, and Koen De Bosschere. 2009. “Factoring Out Ordered Sections to Expose Thread-level Parallelism.” In PEPSMA 2009 : the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures, 12–19.
APA
Vandierendonck, H., Rul, S., & De Bosschere, K. (2009). Factoring out ordered sections to expose thread-level parallelism. PEPSMA 2009 : the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures (pp. 12–19). Presented at the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures (PESPMA 2009).
Vancouver
1.
Vandierendonck H, Rul S, De Bosschere K. Factoring out ordered sections to expose thread-level parallelism. PEPSMA 2009 : the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures. 2009. p. 12–9.
MLA
Vandierendonck, Hans, Sean Rul, and Koen De Bosschere. “Factoring Out Ordered Sections to Expose Thread-level Parallelism.” PEPSMA 2009 : the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures. 2009. 12–19. Print.
@inproceedings{749798,
  abstract     = {With the rise of multi-core processors, researchers are taking a new look at extending the applicability auto-parallelization techniques. In this paper, we identify a dependence pattern on which autoparallelization currently fails. This dependence pattern occurs for ordered sections, i.e. code fragments in a loop that must be executed atomically and in original program order. We discuss why these ordered sections prohibit current auto-parallelizers from working and we present a technique to deal with them. We experimentally demonstrate the efficacy of the technique, yielding significant overall program speedups.},
  author       = {Vandierendonck, Hans and Rul, Sean and De Bosschere, Koen},
  booktitle    = {PEPSMA 2009 : the 2nd Workshop on Parallel Execution of Sequential Programs on Multicore Architectures},
  keyword      = {Multi-core,ordered sections,auto-parallelization},
  language     = {eng},
  location     = {Austin, TX, USA},
  pages        = {12--19},
  title        = {Factoring out ordered sections to expose thread-level parallelism},
  url          = {http://cccp.eecs.umich.edu/pespma/PESPMA-2009-Proceedings.pdf},
  year         = {2009},
}