MTPA: A processor architecture for MP-SOCs employing the moving threads paradigm

Martti Forsell, Ville Leppänen

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    Abstract

    Moving threads is a new kind of approach for mapping the computation of an application to multiprocessor system on chip (MP-SOC) architecture. Instead of moving data read and write requests, extremely lightweight threads are moved between the processor cores. As a consequence, all kinds of cache coherence problems and need for read reply messages are eliminated. Lamport's sequential consistency of shared memory multiprocessor systems is achieved for free. Flexible and efficient handling of varying number of threads is included in the architecture. In this paper we propose a processor architecture (MTPA) for the moving threads paradigm. We describe the overall structure, operation, instruction set, and thread management mechanism as well as evaluate the proposed architecture with different functional unit settings with simulations and give early silicon area and power consumption estimates.
    Original languageEnglish
    Title of host publicationProceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009
    EditorsHamid R. Arabnia
    Pages198-204
    Volume1
    ISBN (Electronic)1-60132-122-8
    Publication statusPublished - 2009
    MoE publication typeA4 Article in a conference publication
    Event2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009 - Las Vegas, United States
    Duration: 13 Jul 200916 Jul 2009

    Conference

    Conference2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009
    Abbreviated titlePDPTA 2009
    CountryUnited States
    CityLas Vegas
    Period13/07/0916/07/09

    Fingerprint

    Electric power utilization
    Data storage equipment
    Silicon
    System-on-chip

    Keywords

    • Moving threads
    • parallel computing
    • MP-SOC
    • processor architecture
    • computer architecture

    Cite this

    Forsell, M., & Leppänen, V. (2009). MTPA: A processor architecture for MP-SOCs employing the moving threads paradigm. In H. R. Arabnia (Ed.), Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009 (Vol. 1, pp. 198-204)
    Forsell, Martti ; Leppänen, Ville. / MTPA : A processor architecture for MP-SOCs employing the moving threads paradigm. Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009. editor / Hamid R. Arabnia. Vol. 1 2009. pp. 198-204
    @inproceedings{39d3357c8a804ea8837053a0d47290b5,
    title = "MTPA: A processor architecture for MP-SOCs employing the moving threads paradigm",
    abstract = "Moving threads is a new kind of approach for mapping the computation of an application to multiprocessor system on chip (MP-SOC) architecture. Instead of moving data read and write requests, extremely lightweight threads are moved between the processor cores. As a consequence, all kinds of cache coherence problems and need for read reply messages are eliminated. Lamport's sequential consistency of shared memory multiprocessor systems is achieved for free. Flexible and efficient handling of varying number of threads is included in the architecture. In this paper we propose a processor architecture (MTPA) for the moving threads paradigm. We describe the overall structure, operation, instruction set, and thread management mechanism as well as evaluate the proposed architecture with different functional unit settings with simulations and give early silicon area and power consumption estimates.",
    keywords = "Moving threads, parallel computing, MP-SOC, processor architecture, computer architecture",
    author = "Martti Forsell and Ville Lepp{\"a}nen",
    year = "2009",
    language = "English",
    isbn = "1-60132-121-X",
    volume = "1",
    pages = "198--204",
    editor = "Arabnia, {Hamid R.}",
    booktitle = "Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009",

    }

    Forsell, M & Leppänen, V 2009, MTPA: A processor architecture for MP-SOCs employing the moving threads paradigm. in HR Arabnia (ed.), Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009. vol. 1, pp. 198-204, 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009, Las Vegas, United States, 13/07/09.

    MTPA : A processor architecture for MP-SOCs employing the moving threads paradigm. / Forsell, Martti; Leppänen, Ville.

    Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009. ed. / Hamid R. Arabnia. Vol. 1 2009. p. 198-204.

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    TY - GEN

    T1 - MTPA

    T2 - A processor architecture for MP-SOCs employing the moving threads paradigm

    AU - Forsell, Martti

    AU - Leppänen, Ville

    PY - 2009

    Y1 - 2009

    N2 - Moving threads is a new kind of approach for mapping the computation of an application to multiprocessor system on chip (MP-SOC) architecture. Instead of moving data read and write requests, extremely lightweight threads are moved between the processor cores. As a consequence, all kinds of cache coherence problems and need for read reply messages are eliminated. Lamport's sequential consistency of shared memory multiprocessor systems is achieved for free. Flexible and efficient handling of varying number of threads is included in the architecture. In this paper we propose a processor architecture (MTPA) for the moving threads paradigm. We describe the overall structure, operation, instruction set, and thread management mechanism as well as evaluate the proposed architecture with different functional unit settings with simulations and give early silicon area and power consumption estimates.

    AB - Moving threads is a new kind of approach for mapping the computation of an application to multiprocessor system on chip (MP-SOC) architecture. Instead of moving data read and write requests, extremely lightweight threads are moved between the processor cores. As a consequence, all kinds of cache coherence problems and need for read reply messages are eliminated. Lamport's sequential consistency of shared memory multiprocessor systems is achieved for free. Flexible and efficient handling of varying number of threads is included in the architecture. In this paper we propose a processor architecture (MTPA) for the moving threads paradigm. We describe the overall structure, operation, instruction set, and thread management mechanism as well as evaluate the proposed architecture with different functional unit settings with simulations and give early silicon area and power consumption estimates.

    KW - Moving threads

    KW - parallel computing

    KW - MP-SOC

    KW - processor architecture

    KW - computer architecture

    M3 - Conference article in proceedings

    SN - 1-60132-121-X

    VL - 1

    SP - 198

    EP - 204

    BT - Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009

    A2 - Arabnia, Hamid R.

    ER -

    Forsell M, Leppänen V. MTPA: A processor architecture for MP-SOCs employing the moving threads paradigm. In Arabnia HR, editor, Proceedings of the 2009 International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009. Vol. 1. 2009. p. 198-204