Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores

Akihiro Hayashi, Yasutaka Wada, Takeshi Watanabe, Takeshi Sekiguchi, Masayoshi Mase, Jun Shirako, Keiji Kimura, Hironori Kasahara

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

    9 Citations (Scopus)

    Abstract

    Heterogeneous multicores have been attracting much attention to attain high performance keeping power consumption low in wide spread of areas. However, heterogeneous multicores force programmers very difficult programming. The long application program development period lowers product competitiveness. In order to overcome such a situation, this paper proposes a compilation framework which bridges a gap between programmers and heterogeneous multicores. In particular, this paper describes the compilation framework based on OSCAR compiler. It realizes coarse grain task parallel processing, data transfer using a DMA controller, power reduction control from user programs with DVFS and clock gating on various heterogeneous multicores from different vendors. This paper also evaluates processing performance and the power reduction by the proposed framework on a newly developed 15 core heterogeneous multicore chip named RP-X integrating 8 general purpose processor cores and 3 types of accelerator cores which was developed by Renesas Electronics, Hitachi, Tokyo Institute of Technology and Waseda University. The framework attains speedups up to 32x for an optical flow program with eight general purpose processor cores and four DRP(Dynamically Reconfigurable Processor) accelerator cores against sequential execution by a single processor core and 80% of power reduction for the real-time AAC encoding.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Pages184-198
    Number of pages15
    Volume6548 LNCS
    DOIs
    Publication statusPublished - 2011
    Event23rd International Workshop on Languages and Compilers for Parallel Computing, LCPC 2010 - Houston, TX
    Duration: 2010 Oct 72010 Oct 9

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume6548 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other23rd International Workshop on Languages and Compilers for Parallel Computing, LCPC 2010
    CityHouston, TX
    Period10/10/710/10/9

    Fingerprint

    Parallelizing Compilers
    Application programming interfaces (API)
    Productivity
    Real-time
    Particle accelerators
    Software
    Compilation
    Accelerator
    Optical flows
    Dynamic mechanical analysis
    Data transfer
    Processing
    Computer programming
    Application programs
    Clocks
    Electric power utilization
    Electronic equipment
    Optical Flow
    Competitiveness
    Data Transfer

    Keywords

    • API
    • Heterogeneous Multicore
    • Parallelizing Compiler

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Hayashi, A., Wada, Y., Watanabe, T., Sekiguchi, T., Mase, M., Shirako, J., ... Kasahara, H. (2011). Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6548 LNCS, pp. 184-198). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6548 LNCS). https://doi.org/10.1007/978-3-642-19595-2_13

    Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores. / Hayashi, Akihiro; Wada, Yasutaka; Watanabe, Takeshi; Sekiguchi, Takeshi; Mase, Masayoshi; Shirako, Jun; Kimura, Keiji; Kasahara, Hironori.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6548 LNCS 2011. p. 184-198 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6548 LNCS).

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

    Hayashi, A, Wada, Y, Watanabe, T, Sekiguchi, T, Mase, M, Shirako, J, Kimura, K & Kasahara, H 2011, Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 6548 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6548 LNCS, pp. 184-198, 23rd International Workshop on Languages and Compilers for Parallel Computing, LCPC 2010, Houston, TX, 10/10/7. https://doi.org/10.1007/978-3-642-19595-2_13
    Hayashi A, Wada Y, Watanabe T, Sekiguchi T, Mase M, Shirako J et al. Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6548 LNCS. 2011. p. 184-198. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-19595-2_13
    Hayashi, Akihiro ; Wada, Yasutaka ; Watanabe, Takeshi ; Sekiguchi, Takeshi ; Mase, Masayoshi ; Shirako, Jun ; Kimura, Keiji ; Kasahara, Hironori. / Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6548 LNCS 2011. pp. 184-198 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{0577d32eb66e43a3a87deca386aa579f,
    title = "Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores",
    abstract = "Heterogeneous multicores have been attracting much attention to attain high performance keeping power consumption low in wide spread of areas. However, heterogeneous multicores force programmers very difficult programming. The long application program development period lowers product competitiveness. In order to overcome such a situation, this paper proposes a compilation framework which bridges a gap between programmers and heterogeneous multicores. In particular, this paper describes the compilation framework based on OSCAR compiler. It realizes coarse grain task parallel processing, data transfer using a DMA controller, power reduction control from user programs with DVFS and clock gating on various heterogeneous multicores from different vendors. This paper also evaluates processing performance and the power reduction by the proposed framework on a newly developed 15 core heterogeneous multicore chip named RP-X integrating 8 general purpose processor cores and 3 types of accelerator cores which was developed by Renesas Electronics, Hitachi, Tokyo Institute of Technology and Waseda University. The framework attains speedups up to 32x for an optical flow program with eight general purpose processor cores and four DRP(Dynamically Reconfigurable Processor) accelerator cores against sequential execution by a single processor core and 80{\%} of power reduction for the real-time AAC encoding.",
    keywords = "API, Heterogeneous Multicore, Parallelizing Compiler",
    author = "Akihiro Hayashi and Yasutaka Wada and Takeshi Watanabe and Takeshi Sekiguchi and Masayoshi Mase and Jun Shirako and Keiji Kimura and Hironori Kasahara",
    year = "2011",
    doi = "10.1007/978-3-642-19595-2_13",
    language = "English",
    isbn = "9783642195945",
    volume = "6548 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "184--198",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - Parallelizing compiler framework and API for power reduction and software productivity of real-time heterogeneous multicores

    AU - Hayashi, Akihiro

    AU - Wada, Yasutaka

    AU - Watanabe, Takeshi

    AU - Sekiguchi, Takeshi

    AU - Mase, Masayoshi

    AU - Shirako, Jun

    AU - Kimura, Keiji

    AU - Kasahara, Hironori

    PY - 2011

    Y1 - 2011

    N2 - Heterogeneous multicores have been attracting much attention to attain high performance keeping power consumption low in wide spread of areas. However, heterogeneous multicores force programmers very difficult programming. The long application program development period lowers product competitiveness. In order to overcome such a situation, this paper proposes a compilation framework which bridges a gap between programmers and heterogeneous multicores. In particular, this paper describes the compilation framework based on OSCAR compiler. It realizes coarse grain task parallel processing, data transfer using a DMA controller, power reduction control from user programs with DVFS and clock gating on various heterogeneous multicores from different vendors. This paper also evaluates processing performance and the power reduction by the proposed framework on a newly developed 15 core heterogeneous multicore chip named RP-X integrating 8 general purpose processor cores and 3 types of accelerator cores which was developed by Renesas Electronics, Hitachi, Tokyo Institute of Technology and Waseda University. The framework attains speedups up to 32x for an optical flow program with eight general purpose processor cores and four DRP(Dynamically Reconfigurable Processor) accelerator cores against sequential execution by a single processor core and 80% of power reduction for the real-time AAC encoding.

    AB - Heterogeneous multicores have been attracting much attention to attain high performance keeping power consumption low in wide spread of areas. However, heterogeneous multicores force programmers very difficult programming. The long application program development period lowers product competitiveness. In order to overcome such a situation, this paper proposes a compilation framework which bridges a gap between programmers and heterogeneous multicores. In particular, this paper describes the compilation framework based on OSCAR compiler. It realizes coarse grain task parallel processing, data transfer using a DMA controller, power reduction control from user programs with DVFS and clock gating on various heterogeneous multicores from different vendors. This paper also evaluates processing performance and the power reduction by the proposed framework on a newly developed 15 core heterogeneous multicore chip named RP-X integrating 8 general purpose processor cores and 3 types of accelerator cores which was developed by Renesas Electronics, Hitachi, Tokyo Institute of Technology and Waseda University. The framework attains speedups up to 32x for an optical flow program with eight general purpose processor cores and four DRP(Dynamically Reconfigurable Processor) accelerator cores against sequential execution by a single processor core and 80% of power reduction for the real-time AAC encoding.

    KW - API

    KW - Heterogeneous Multicore

    KW - Parallelizing Compiler

    UR - http://www.scopus.com/inward/record.url?scp=79952612610&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=79952612610&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-19595-2_13

    DO - 10.1007/978-3-642-19595-2_13

    M3 - Conference contribution

    AN - SCOPUS:79952612610

    SN - 9783642195945

    VL - 6548 LNCS

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 184

    EP - 198

    BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    ER -