Advanced search
1 file | 477.72 KB Add to list

A case study in coordination programming: performance evaluation of S-Net vs. Intel's concurrent collections

Author
Organization
Abstract
We present a programming methodology and runtime performance case study comparing the declarative data flow coordination language S-Net with Intel's Concurrent Collections (CnC). As a coordination language S-Net achieves a near-complete separation of concerns between sequential software components implemented in a separate algorithmic language and their parallel orchestration in an asynchronous data flow streaming network. We investigate the merits of S-Net and CnC with the help of a relevant and non-trivial linear algebra problem: tiled Cholesky decomposition. We describe two alternative S-Net implementations of tiled Cholesky factorization and compare them with two CnC implementations, one with explicit performance tuning and one without, that have previously been used to illustrate Intel CnC. Our experiments on a 48-core machine demonstrate that S-Net manages to outperform CnC on this problem.
Keywords
coordination programming, performance measurement, stream processing, concurrent collections, parallel programming, language design

Downloads

  • IPDPS 2014 cholesky.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 477.72 KB

Citation

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

MLA
Zaichenkov, Pavel, et al. “A Case Study in Coordination Programming: Performance Evaluation of S-Net vs. Intel’s Concurrent Collections.” Parallel and Distributed Processing Symposium Workshops, 28th IEEE International, IEEE, 2014, pp. 1059–67, doi:10.1109/IPDPSW.2014.118.
APA
Zaichenkov, P., Gijsbers, B., Grelck, C., Tveretina, O., & Shafarenko, A. (2014). A case study in coordination programming: performance evaluation of S-Net vs. Intel’s concurrent collections. In Parallel and Distributed Processing Symposium Workshops, 28th IEEE International (pp. 1059–1067). New York, NY, USA: IEEE. https://doi.org/10.1109/IPDPSW.2014.118
Chicago author-date
Zaichenkov, Pavel, Bert Gijsbers, Clemens Grelck, Olga Tveretina, and Alex Shafarenko. 2014. “A Case Study in Coordination Programming: Performance Evaluation of S-Net vs. Intel’s Concurrent Collections.” In Parallel and Distributed Processing Symposium Workshops, 28th IEEE International, 1059–67. New York, NY, USA: IEEE. https://doi.org/10.1109/IPDPSW.2014.118.
Chicago author-date (all authors)
Zaichenkov, Pavel, Bert Gijsbers, Clemens Grelck, Olga Tveretina, and Alex Shafarenko. 2014. “A Case Study in Coordination Programming: Performance Evaluation of S-Net vs. Intel’s Concurrent Collections.” In Parallel and Distributed Processing Symposium Workshops, 28th IEEE International, 1059–1067. New York, NY, USA: IEEE. doi:10.1109/IPDPSW.2014.118.
Vancouver
1.
Zaichenkov P, Gijsbers B, Grelck C, Tveretina O, Shafarenko A. A case study in coordination programming: performance evaluation of S-Net vs. Intel’s concurrent collections. In: Parallel and Distributed Processing Symposium Workshops, 28th IEEE International. New York, NY, USA: IEEE; 2014. p. 1059–67.
IEEE
[1]
P. Zaichenkov, B. Gijsbers, C. Grelck, O. Tveretina, and A. Shafarenko, “A case study in coordination programming: performance evaluation of S-Net vs. Intel’s concurrent collections,” in Parallel and Distributed Processing Symposium Workshops, 28th IEEE International, Phoenix, AZ, USA, 2014, pp. 1059–1067.
@inproceedings{4358180,
  abstract     = {{We present a programming methodology and runtime performance case study comparing the declarative data flow coordination language S-Net with Intel's Concurrent Collections (CnC). As a coordination language S-Net achieves a near-complete separation of concerns between sequential software components implemented in a separate algorithmic language and their parallel orchestration in an asynchronous data flow streaming network. We investigate the merits of S-Net and CnC with the help of a relevant and non-trivial linear algebra problem: tiled Cholesky decomposition. We describe two alternative S-Net implementations of tiled Cholesky factorization and compare them with two CnC implementations, one with explicit performance tuning and one without, that have previously been used to illustrate Intel CnC. Our experiments on a 48-core machine demonstrate that S-Net manages to outperform CnC on this problem.}},
  author       = {{Zaichenkov, Pavel and Gijsbers, Bert and Grelck, Clemens and Tveretina, Olga and Shafarenko, Alex}},
  booktitle    = {{Parallel and Distributed Processing Symposium Workshops, 28th IEEE International}},
  isbn         = {{9781479941179}},
  keywords     = {{coordination programming,performance measurement,stream processing,concurrent collections,parallel programming,language design}},
  language     = {{eng}},
  location     = {{Phoenix, AZ, USA}},
  pages        = {{1059--1067}},
  publisher    = {{IEEE}},
  title        = {{A case study in coordination programming: performance evaluation of S-Net vs. Intel's concurrent collections}},
  url          = {{http://dx.doi.org/10.1109/IPDPSW.2014.118}},
  year         = {{2014}},
}

Altmetric
View in Altmetric