Multigrain parallel processing on compiler cooperative chip multiprocessor

Keiji Kimura, Yasutaka Wada, Hirufumi Nakano, Takeshi Kodaka, Jun Shirako, Kazuhisa Ishizaka, Hironori Kasahara

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

    13 Citations (Scopus)

    Abstract

    This paper describes multigrain parallel processing on a compiler cooperative chip multiprocessor. The multigrain parallel processing hierarchically exploits multiple grains of parallelism such as coarse grain task parallelism, loop iteration level parallelism and statement level near-fine grain parallelism. The chip multiprocessor has been designed to attain high effective performance, cost effectiveness and high software productivity by supporting the optimizations of the multigrain parallelizing compiler, which is developed by Japanese Millennium Project IT21 "Advance Parallelizing Compiler". To achieve full potential of multigrain parallel processing, the chip multiprocessor integrates simple single-issue processors having distributed shared data memory for both optimal use of data locality and scalar data transfer, local data memory for processor private data, in addition to centralized shared memory for shared data among processors. This paper focuses on the scalability of the chip multiprocessor having up to eight processors on a chip by exploiting of the multigrain parallelism from SPECfp95 programs. When microSPARC like the simple processor core is used under assumption of 90 nm technology and 2.8 GHz, the evaluation results show the speedups for eight processors and four processors reach 7.1 and 3.9, respectively. Similarly, when 400 MHz is assumed for embedded usage, the speedups reach 7.8 and 4.0, respectively.

    Original languageEnglish
    Title of host publicationProceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT
    Pages11-21
    Number of pages11
    Volume2005
    DOIs
    Publication statusPublished - 2005
    Event9th Annual Workshop on Interaction between Compilers and Computer Architectures INTERACT-9, in conjunction with the 11th International Symposium on High-performance Computer Architecture, HPCA-11 - San Francisco, CA, United States
    Duration: 2005 Feb 132005 Feb 13

    Other

    Other9th Annual Workshop on Interaction between Compilers and Computer Architectures INTERACT-9, in conjunction with the 11th International Symposium on High-performance Computer Architecture, HPCA-11
    CountryUnited States
    CitySan Francisco, CA
    Period05/2/1305/2/13

    Fingerprint

    Data storage equipment
    Processing
    Data transfer
    Cost effectiveness
    Scalability
    Productivity

    ASJC Scopus subject areas

    • Engineering(all)

    Cite this

    Kimura, K., Wada, Y., Nakano, H., Kodaka, T., Shirako, J., Ishizaka, K., & Kasahara, H. (2005). Multigrain parallel processing on compiler cooperative chip multiprocessor. In Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT (Vol. 2005, pp. 11-21). [1423137] https://doi.org/10.1109/INTERACT.2005.9

    Multigrain parallel processing on compiler cooperative chip multiprocessor. / Kimura, Keiji; Wada, Yasutaka; Nakano, Hirufumi; Kodaka, Takeshi; Shirako, Jun; Ishizaka, Kazuhisa; Kasahara, Hironori.

    Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT. Vol. 2005 2005. p. 11-21 1423137.

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

    Kimura, K, Wada, Y, Nakano, H, Kodaka, T, Shirako, J, Ishizaka, K & Kasahara, H 2005, Multigrain parallel processing on compiler cooperative chip multiprocessor. in Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT. vol. 2005, 1423137, pp. 11-21, 9th Annual Workshop on Interaction between Compilers and Computer Architectures INTERACT-9, in conjunction with the 11th International Symposium on High-performance Computer Architecture, HPCA-11, San Francisco, CA, United States, 05/2/13. https://doi.org/10.1109/INTERACT.2005.9
    Kimura K, Wada Y, Nakano H, Kodaka T, Shirako J, Ishizaka K et al. Multigrain parallel processing on compiler cooperative chip multiprocessor. In Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT. Vol. 2005. 2005. p. 11-21. 1423137 https://doi.org/10.1109/INTERACT.2005.9
    Kimura, Keiji ; Wada, Yasutaka ; Nakano, Hirufumi ; Kodaka, Takeshi ; Shirako, Jun ; Ishizaka, Kazuhisa ; Kasahara, Hironori. / Multigrain parallel processing on compiler cooperative chip multiprocessor. Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT. Vol. 2005 2005. pp. 11-21
    @inproceedings{ba9974e3e8004cb4a7bbaed8d55a2288,
    title = "Multigrain parallel processing on compiler cooperative chip multiprocessor",
    abstract = "This paper describes multigrain parallel processing on a compiler cooperative chip multiprocessor. The multigrain parallel processing hierarchically exploits multiple grains of parallelism such as coarse grain task parallelism, loop iteration level parallelism and statement level near-fine grain parallelism. The chip multiprocessor has been designed to attain high effective performance, cost effectiveness and high software productivity by supporting the optimizations of the multigrain parallelizing compiler, which is developed by Japanese Millennium Project IT21 {"}Advance Parallelizing Compiler{"}. To achieve full potential of multigrain parallel processing, the chip multiprocessor integrates simple single-issue processors having distributed shared data memory for both optimal use of data locality and scalar data transfer, local data memory for processor private data, in addition to centralized shared memory for shared data among processors. This paper focuses on the scalability of the chip multiprocessor having up to eight processors on a chip by exploiting of the multigrain parallelism from SPECfp95 programs. When microSPARC like the simple processor core is used under assumption of 90 nm technology and 2.8 GHz, the evaluation results show the speedups for eight processors and four processors reach 7.1 and 3.9, respectively. Similarly, when 400 MHz is assumed for embedded usage, the speedups reach 7.8 and 4.0, respectively.",
    author = "Keiji Kimura and Yasutaka Wada and Hirufumi Nakano and Takeshi Kodaka and Jun Shirako and Kazuhisa Ishizaka and Hironori Kasahara",
    year = "2005",
    doi = "10.1109/INTERACT.2005.9",
    language = "English",
    volume = "2005",
    pages = "11--21",
    booktitle = "Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT",

    }

    TY - GEN

    T1 - Multigrain parallel processing on compiler cooperative chip multiprocessor

    AU - Kimura, Keiji

    AU - Wada, Yasutaka

    AU - Nakano, Hirufumi

    AU - Kodaka, Takeshi

    AU - Shirako, Jun

    AU - Ishizaka, Kazuhisa

    AU - Kasahara, Hironori

    PY - 2005

    Y1 - 2005

    N2 - This paper describes multigrain parallel processing on a compiler cooperative chip multiprocessor. The multigrain parallel processing hierarchically exploits multiple grains of parallelism such as coarse grain task parallelism, loop iteration level parallelism and statement level near-fine grain parallelism. The chip multiprocessor has been designed to attain high effective performance, cost effectiveness and high software productivity by supporting the optimizations of the multigrain parallelizing compiler, which is developed by Japanese Millennium Project IT21 "Advance Parallelizing Compiler". To achieve full potential of multigrain parallel processing, the chip multiprocessor integrates simple single-issue processors having distributed shared data memory for both optimal use of data locality and scalar data transfer, local data memory for processor private data, in addition to centralized shared memory for shared data among processors. This paper focuses on the scalability of the chip multiprocessor having up to eight processors on a chip by exploiting of the multigrain parallelism from SPECfp95 programs. When microSPARC like the simple processor core is used under assumption of 90 nm technology and 2.8 GHz, the evaluation results show the speedups for eight processors and four processors reach 7.1 and 3.9, respectively. Similarly, when 400 MHz is assumed for embedded usage, the speedups reach 7.8 and 4.0, respectively.

    AB - This paper describes multigrain parallel processing on a compiler cooperative chip multiprocessor. The multigrain parallel processing hierarchically exploits multiple grains of parallelism such as coarse grain task parallelism, loop iteration level parallelism and statement level near-fine grain parallelism. The chip multiprocessor has been designed to attain high effective performance, cost effectiveness and high software productivity by supporting the optimizations of the multigrain parallelizing compiler, which is developed by Japanese Millennium Project IT21 "Advance Parallelizing Compiler". To achieve full potential of multigrain parallel processing, the chip multiprocessor integrates simple single-issue processors having distributed shared data memory for both optimal use of data locality and scalar data transfer, local data memory for processor private data, in addition to centralized shared memory for shared data among processors. This paper focuses on the scalability of the chip multiprocessor having up to eight processors on a chip by exploiting of the multigrain parallelism from SPECfp95 programs. When microSPARC like the simple processor core is used under assumption of 90 nm technology and 2.8 GHz, the evaluation results show the speedups for eight processors and four processors reach 7.1 and 3.9, respectively. Similarly, when 400 MHz is assumed for embedded usage, the speedups reach 7.8 and 4.0, respectively.

    UR - http://www.scopus.com/inward/record.url?scp=33744457984&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33744457984&partnerID=8YFLogxK

    U2 - 10.1109/INTERACT.2005.9

    DO - 10.1109/INTERACT.2005.9

    M3 - Conference contribution

    AN - SCOPUS:33744457984

    VL - 2005

    SP - 11

    EP - 21

    BT - Proceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT

    ER -