Clustering-Based Task Scheduling in a Large Number of Heterogeneous Processors

Hidehiro Kanemitsu, Masaki Hanada, Hidenori Nakazato

    Research output: Contribution to journalArticle

    25 Citations (Scopus)

    Abstract

    Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.

    Original languageEnglish
    Article number7401062
    Pages (from-to)3144-3157
    Number of pages14
    JournalIEEE Transactions on Parallel and Distributed Systems
    Volume27
    Issue number11
    DOIs
    Publication statusPublished - 2016 Nov 1

    Fingerprint

    Scheduling algorithms
    Scheduling
    Bandwidth
    Communication

    Keywords

    • DAG scheduling
    • heterogeneous systems
    • task clustering
    • Task scheduling

    ASJC Scopus subject areas

    • Signal Processing
    • Hardware and Architecture
    • Computational Theory and Mathematics

    Cite this

    Clustering-Based Task Scheduling in a Large Number of Heterogeneous Processors. / Kanemitsu, Hidehiro; Hanada, Masaki; Nakazato, Hidenori.

    In: IEEE Transactions on Parallel and Distributed Systems, Vol. 27, No. 11, 7401062, 01.11.2016, p. 3144-3157.

    Research output: Contribution to journalArticle

    @article{5d67ed9265a94022873a8520127eb3dd,
    title = "Clustering-Based Task Scheduling in a Large Number of Heterogeneous Processors",
    abstract = "Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.",
    keywords = "DAG scheduling, heterogeneous systems, task clustering, Task scheduling",
    author = "Hidehiro Kanemitsu and Masaki Hanada and Hidenori Nakazato",
    year = "2016",
    month = "11",
    day = "1",
    doi = "10.1109/TPDS.2016.2526682",
    language = "English",
    volume = "27",
    pages = "3144--3157",
    journal = "IEEE Transactions on Parallel and Distributed Systems",
    issn = "1045-9219",
    publisher = "IEEE Computer Society",
    number = "11",

    }

    TY - JOUR

    T1 - Clustering-Based Task Scheduling in a Large Number of Heterogeneous Processors

    AU - Kanemitsu, Hidehiro

    AU - Hanada, Masaki

    AU - Nakazato, Hidenori

    PY - 2016/11/1

    Y1 - 2016/11/1

    N2 - Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.

    AB - Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.

    KW - DAG scheduling

    KW - heterogeneous systems

    KW - task clustering

    KW - Task scheduling

    UR - http://www.scopus.com/inward/record.url?scp=84994476443&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84994476443&partnerID=8YFLogxK

    U2 - 10.1109/TPDS.2016.2526682

    DO - 10.1109/TPDS.2016.2526682

    M3 - Article

    AN - SCOPUS:84994476443

    VL - 27

    SP - 3144

    EP - 3157

    JO - IEEE Transactions on Parallel and Distributed Systems

    JF - IEEE Transactions on Parallel and Distributed Systems

    SN - 1045-9219

    IS - 11

    M1 - 7401062

    ER -