S-Net for multi-memory multicores

C. Grelck, Jukka Julku, F. Penczek

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

    1 Citation (Scopus)

    Abstract

    S-NET is a declarative coordination language and component technology aimed at modern multi-core/many-core architectures and systems-on-chip. It builds on the concept of stream processing to structure dynamically evolving networks of communicating asynchronous components. Components themselves are implemented using a conventional language suitable for the application domain. This two-level software architecture maintains a familiar sequential development environment for large parts of an application and offers a high-level declarative approach to component coordination. In this paper we present a conservative language extension for the placement of components and component networks in a multi-memory environment, i.e. architectures that associate individual compute cores or groups thereof with private memories. We describe a novel distributed runtime system layer that complements our existing multithreaded runtime system for shared memory multicores. Particular emphasis is put on efficient management of data communication. Last not least, we present preliminary experimental data.
    Original languageEnglish
    Title of host publicationDAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming
    Subtitle of host publicationMadrid, Spain — January 19, 2010
    PublisherAssociation for Computing Machinery ACM
    Pages25-34
    ISBN (Print)978-1-60558-859-9
    DOIs
    Publication statusPublished - 2010
    MoE publication typeA4 Article in a conference publication

    Fingerprint

    Data storage equipment
    Network components
    Software architecture
    Communication
    Processing
    System-on-chip

    Keywords

    • Component coordination
    • Hybrid memory architecture
    • Multicore programming
    • Stream processing

    Cite this

    Grelck, C., Julku, J., & Penczek, F. (2010). S-Net for multi-memory multicores. In DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming: Madrid, Spain — January 19, 2010 (pp. 25-34). Association for Computing Machinery ACM. https://doi.org/10.1145/1708046.1708054
    Grelck, C. ; Julku, Jukka ; Penczek, F. / S-Net for multi-memory multicores. DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming: Madrid, Spain — January 19, 2010. Association for Computing Machinery ACM, 2010. pp. 25-34
    @inproceedings{c5afffd609a744f58e9c1082d67bf788,
    title = "S-Net for multi-memory multicores",
    abstract = "S-NET is a declarative coordination language and component technology aimed at modern multi-core/many-core architectures and systems-on-chip. It builds on the concept of stream processing to structure dynamically evolving networks of communicating asynchronous components. Components themselves are implemented using a conventional language suitable for the application domain. This two-level software architecture maintains a familiar sequential development environment for large parts of an application and offers a high-level declarative approach to component coordination. In this paper we present a conservative language extension for the placement of components and component networks in a multi-memory environment, i.e. architectures that associate individual compute cores or groups thereof with private memories. We describe a novel distributed runtime system layer that complements our existing multithreaded runtime system for shared memory multicores. Particular emphasis is put on efficient management of data communication. Last not least, we present preliminary experimental data.",
    keywords = "Component coordination, Hybrid memory architecture, Multicore programming, Stream processing",
    author = "C. Grelck and Jukka Julku and F. Penczek",
    year = "2010",
    doi = "10.1145/1708046.1708054",
    language = "English",
    isbn = "978-1-60558-859-9",
    pages = "25--34",
    booktitle = "DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming",
    publisher = "Association for Computing Machinery ACM",
    address = "United States",

    }

    Grelck, C, Julku, J & Penczek, F 2010, S-Net for multi-memory multicores. in DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming: Madrid, Spain — January 19, 2010. Association for Computing Machinery ACM, pp. 25-34. https://doi.org/10.1145/1708046.1708054

    S-Net for multi-memory multicores. / Grelck, C.; Julku, Jukka; Penczek, F.

    DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming: Madrid, Spain — January 19, 2010. Association for Computing Machinery ACM, 2010. p. 25-34.

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

    TY - GEN

    T1 - S-Net for multi-memory multicores

    AU - Grelck, C.

    AU - Julku, Jukka

    AU - Penczek, F.

    PY - 2010

    Y1 - 2010

    N2 - S-NET is a declarative coordination language and component technology aimed at modern multi-core/many-core architectures and systems-on-chip. It builds on the concept of stream processing to structure dynamically evolving networks of communicating asynchronous components. Components themselves are implemented using a conventional language suitable for the application domain. This two-level software architecture maintains a familiar sequential development environment for large parts of an application and offers a high-level declarative approach to component coordination. In this paper we present a conservative language extension for the placement of components and component networks in a multi-memory environment, i.e. architectures that associate individual compute cores or groups thereof with private memories. We describe a novel distributed runtime system layer that complements our existing multithreaded runtime system for shared memory multicores. Particular emphasis is put on efficient management of data communication. Last not least, we present preliminary experimental data.

    AB - S-NET is a declarative coordination language and component technology aimed at modern multi-core/many-core architectures and systems-on-chip. It builds on the concept of stream processing to structure dynamically evolving networks of communicating asynchronous components. Components themselves are implemented using a conventional language suitable for the application domain. This two-level software architecture maintains a familiar sequential development environment for large parts of an application and offers a high-level declarative approach to component coordination. In this paper we present a conservative language extension for the placement of components and component networks in a multi-memory environment, i.e. architectures that associate individual compute cores or groups thereof with private memories. We describe a novel distributed runtime system layer that complements our existing multithreaded runtime system for shared memory multicores. Particular emphasis is put on efficient management of data communication. Last not least, we present preliminary experimental data.

    KW - Component coordination

    KW - Hybrid memory architecture

    KW - Multicore programming

    KW - Stream processing

    U2 - 10.1145/1708046.1708054

    DO - 10.1145/1708046.1708054

    M3 - Conference article in proceedings

    SN - 978-1-60558-859-9

    SP - 25

    EP - 34

    BT - DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming

    PB - Association for Computing Machinery ACM

    ER -

    Grelck C, Julku J, Penczek F. S-Net for multi-memory multicores. In DAMP'10 Proceedings of the 2010 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming: Madrid, Spain — January 19, 2010. Association for Computing Machinery ACM. 2010. p. 25-34 https://doi.org/10.1145/1708046.1708054