Performance analysis and linear optimization modeling of all-to-all collective communication algorithms

Hyacinthe Nzigou Mamadou, Guilherme De Melo Baptista Domingues, Takeshi Nanri, Kazuaki Murakami

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    2 Citations (Scopus)

    Abstract

    The performance of collective communication operations still represents a critical issue for high performance computing systems. Users of parallel machines need to have a good grasp of how different communication patterns and styles affect the performance of message-passing applications. This paper reports our contribution of the analysis of collective communication algorithms in the context of MPI programming paradigm by extending a standard pointto-point communication model, which is P-LogP. We focus on MPI Alltoall since this function is one of the most communication intensive collective operations known. In order to reduce the gap between the predicted and the measured run-time, all the system parameters are also taken into account with the total performance estimation, by applying the linear regression modeling with the empirical data. Results on InfiniBand clusters show that the final performance prediction models can accurately capture the entire system communication behavior of all algorithms, even for large size messages and large number of processors.

    Original languageEnglish
    Title of host publicationProceedings - 19th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD
    Pages203-210
    Number of pages8
    DOIs
    Publication statusPublished - 2007
    Event19th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD - Gramado, RS, Brazil
    Duration: Oct 24 2007Oct 27 2007

    Publication series

    NameProceedings - Symposium on Computer Architecture and High Performance Computing
    ISSN (Print)1550-6533

    Other

    Other19th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD
    Country/TerritoryBrazil
    CityGramado, RS
    Period10/24/0710/27/07

    All Science Journal Classification (ASJC) codes

    • Engineering(all)

    Fingerprint

    Dive into the research topics of 'Performance analysis and linear optimization modeling of all-to-all collective communication algorithms'. Together they form a unique fingerprint.

    Cite this