article · 2002

A comparison of data-parallel collective communication performance and its application

Yoshio Tanaka, Kazuki Kubota, Motohiko Matsuda, Mitsuhisa Sato, Satoshi Sekiguchi

Cited by 1Abstract and details only

Abstract

Collective communications such as broadcast and reduction are commonly used in data parallel programs. It is important to understand the performance of such primitive communications to characterize parallel systems and analyze the performance of parallel applications running on specific parallel systems. We measured the performance of collective communication operations on several multiprocessor systems. In this paper, we report experimental results for collective communication performance on distributed memory systems. We also describe the performance prediction of data parallel programs using the performance of the primitives.

Cite this paper

APA 7

Tanaka, Y., Kubota, K., Matsuda, M., Sato, M., & Sekiguchi, S. (2002). A comparison of data-parallel collective communication performance and its application. https://doi.org/10.1109/hpc.1997.592137

MLA 9

Tanaka, Yoshio, et al. “A comparison of data-parallel collective communication performance and its application.” vol. 96, 2002, pp. 137–44, https://doi.org/10.1109/hpc.1997.592137.

Chicago

Tanaka, Yoshio, Kazuki Kubota, Motohiko Matsuda, Mitsuhisa Sato, and Satoshi Sekiguchi. 2002. “A comparison of data-parallel collective communication performance and its application.” https://doi.org/10.1109/hpc.1997.592137.

Harvard

Tanaka, Y. et al. (2002) ‘A comparison of data-parallel collective communication performance and its application’. Available at: https://doi.org/10.1109/hpc.1997.592137.

BibTeX

@article{tanaka2002comparison,
  title = {{A comparison of data-parallel collective communication performance and its application}},
  author = {Tanaka, Yoshio and Kubota, Kazuki and Matsuda, Motohiko and Sato, Mitsuhisa and Sekiguchi, Satoshi},
  volume = {96},
  pages = {137--144},
  year = {2002},
  doi = {10.1109/hpc.1997.592137}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.