Data-localization scheduling inside processor-cluster for multigrain parallel processing

Akimasa Yoshida, KeN'Ichi Koshizuka, Wataru Ogata, Hironori Kasahara

    Research output: Contribution to journalArticle

    Abstract

    This paper proposes a data-localization scheduling scheme inside a processor-cluster for multigrain parallel processing, which hierarchically exploits parallelism among coarsegrain tasks like loops, medium-grain tasks like loop iterations and near-fine-grain tasks like statements. The proposed scheme assigns near-fine-grain or medium-grain tasks inside coarse-grain tasks onto processors inside a processor-cluster so that maximum parallelism can be exploited and inter-processor data transfer can be minimum after data-localization for coarse-grain tasks across processor-clusters. Performance evaluation on a multiprocessor system OSCAR shows that multigrain parallel processing with the proposed data-localization scheduling can reduce execution time for application programs by 10% compared with multigrain parallel processing without data-localization.

    Original languageEnglish
    Pages (from-to)473-478
    Number of pages6
    JournalIEICE Transactions on Information and Systems
    VolumeE80-D
    Issue number4
    Publication statusPublished - 1997

    Fingerprint

    Scheduling
    Processing
    Data transfer
    Application programs

    Keywords

    • Automatic data decomposition
    • Data-localization
    • Multigrain parallel processing
    • Parallelizing compilers
    • Task scheduling

    ASJC Scopus subject areas

    • Information Systems
    • Computer Graphics and Computer-Aided Design
    • Software

    Cite this

    Data-localization scheduling inside processor-cluster for multigrain parallel processing. / Yoshida, Akimasa; Koshizuka, KeN'Ichi; Ogata, Wataru; Kasahara, Hironori.

    In: IEICE Transactions on Information and Systems, Vol. E80-D, No. 4, 1997, p. 473-478.

    Research output: Contribution to journalArticle

    Yoshida, Akimasa ; Koshizuka, KeN'Ichi ; Ogata, Wataru ; Kasahara, Hironori. / Data-localization scheduling inside processor-cluster for multigrain parallel processing. In: IEICE Transactions on Information and Systems. 1997 ; Vol. E80-D, No. 4. pp. 473-478.
    @article{1b587a0ab5ae47f7a4d44da11f7871cc,
    title = "Data-localization scheduling inside processor-cluster for multigrain parallel processing",
    abstract = "This paper proposes a data-localization scheduling scheme inside a processor-cluster for multigrain parallel processing, which hierarchically exploits parallelism among coarsegrain tasks like loops, medium-grain tasks like loop iterations and near-fine-grain tasks like statements. The proposed scheme assigns near-fine-grain or medium-grain tasks inside coarse-grain tasks onto processors inside a processor-cluster so that maximum parallelism can be exploited and inter-processor data transfer can be minimum after data-localization for coarse-grain tasks across processor-clusters. Performance evaluation on a multiprocessor system OSCAR shows that multigrain parallel processing with the proposed data-localization scheduling can reduce execution time for application programs by 10{\%} compared with multigrain parallel processing without data-localization.",
    keywords = "Automatic data decomposition, Data-localization, Multigrain parallel processing, Parallelizing compilers, Task scheduling",
    author = "Akimasa Yoshida and KeN'Ichi Koshizuka and Wataru Ogata and Hironori Kasahara",
    year = "1997",
    language = "English",
    volume = "E80-D",
    pages = "473--478",
    journal = "IEICE Transactions on Information and Systems",
    issn = "0916-8532",
    publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
    number = "4",

    }

    TY - JOUR

    T1 - Data-localization scheduling inside processor-cluster for multigrain parallel processing

    AU - Yoshida, Akimasa

    AU - Koshizuka, KeN'Ichi

    AU - Ogata, Wataru

    AU - Kasahara, Hironori

    PY - 1997

    Y1 - 1997

    N2 - This paper proposes a data-localization scheduling scheme inside a processor-cluster for multigrain parallel processing, which hierarchically exploits parallelism among coarsegrain tasks like loops, medium-grain tasks like loop iterations and near-fine-grain tasks like statements. The proposed scheme assigns near-fine-grain or medium-grain tasks inside coarse-grain tasks onto processors inside a processor-cluster so that maximum parallelism can be exploited and inter-processor data transfer can be minimum after data-localization for coarse-grain tasks across processor-clusters. Performance evaluation on a multiprocessor system OSCAR shows that multigrain parallel processing with the proposed data-localization scheduling can reduce execution time for application programs by 10% compared with multigrain parallel processing without data-localization.

    AB - This paper proposes a data-localization scheduling scheme inside a processor-cluster for multigrain parallel processing, which hierarchically exploits parallelism among coarsegrain tasks like loops, medium-grain tasks like loop iterations and near-fine-grain tasks like statements. The proposed scheme assigns near-fine-grain or medium-grain tasks inside coarse-grain tasks onto processors inside a processor-cluster so that maximum parallelism can be exploited and inter-processor data transfer can be minimum after data-localization for coarse-grain tasks across processor-clusters. Performance evaluation on a multiprocessor system OSCAR shows that multigrain parallel processing with the proposed data-localization scheduling can reduce execution time for application programs by 10% compared with multigrain parallel processing without data-localization.

    KW - Automatic data decomposition

    KW - Data-localization

    KW - Multigrain parallel processing

    KW - Parallelizing compilers

    KW - Task scheduling

    UR - http://www.scopus.com/inward/record.url?scp=0031121930&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0031121930&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:0031121930

    VL - E80-D

    SP - 473

    EP - 478

    JO - IEICE Transactions on Information and Systems

    JF - IEICE Transactions on Information and Systems

    SN - 0916-8532

    IS - 4

    ER -