A data intensive heuristic approach to the two-stage streaming scheduling problem

Wei Liang, Chunhua Hu, Min Wu, Qun Jin

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    Data intensive computing (DIC) provides a high performance computing approach to process large volume of data. In this study, a new formalization is introduced to present the two-stage DIC task execution in a stream manner. A novel heuristic algorithm is proposed for the scheduling problem due to the NP complexity. The theoretical approximation ratio bounds for the heuristic are analyzed and confirmed by the experimental evaluation. Overall, we observe that the proposed method conducts average 1.2 times makespan than the theoretic bound of the optimal solution. Besides, the proposed method outperforms the GA and FIFO scheduling schemes with overall improvements.

    Original languageEnglish
    JournalJournal of Computer and System Sciences
    DOIs
    Publication statusAccepted/In press - 2016 Feb 27

    Fingerprint

    Streaming
    Scheduling Problem
    Scheduling
    Heuristics
    Computing
    Heuristic algorithms
    Time-average
    Formalization
    Experimental Evaluation
    Heuristic algorithm
    High Performance
    Optimal Solution
    Approximation

    Keywords

    • Data intensive computing
    • Makespan
    • NP-hard
    • Scheduling

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Networks and Communications
    • Computational Theory and Mathematics
    • Applied Mathematics

    Cite this

    A data intensive heuristic approach to the two-stage streaming scheduling problem. / Liang, Wei; Hu, Chunhua; Wu, Min; Jin, Qun.

    In: Journal of Computer and System Sciences, 27.02.2016.

    Research output: Contribution to journalArticle

    @article{838dc5845a804633af5fdcdb671882ca,
    title = "A data intensive heuristic approach to the two-stage streaming scheduling problem",
    abstract = "Data intensive computing (DIC) provides a high performance computing approach to process large volume of data. In this study, a new formalization is introduced to present the two-stage DIC task execution in a stream manner. A novel heuristic algorithm is proposed for the scheduling problem due to the NP complexity. The theoretical approximation ratio bounds for the heuristic are analyzed and confirmed by the experimental evaluation. Overall, we observe that the proposed method conducts average 1.2 times makespan than the theoretic bound of the optimal solution. Besides, the proposed method outperforms the GA and FIFO scheduling schemes with overall improvements.",
    keywords = "Data intensive computing, Makespan, NP-hard, Scheduling",
    author = "Wei Liang and Chunhua Hu and Min Wu and Qun Jin",
    year = "2016",
    month = "2",
    day = "27",
    doi = "10.1016/j.jcss.2017.01.005",
    language = "English",
    journal = "Journal of Computer and System Sciences",
    issn = "0022-0000",
    publisher = "Academic Press Inc.",

    }

    TY - JOUR

    T1 - A data intensive heuristic approach to the two-stage streaming scheduling problem

    AU - Liang, Wei

    AU - Hu, Chunhua

    AU - Wu, Min

    AU - Jin, Qun

    PY - 2016/2/27

    Y1 - 2016/2/27

    N2 - Data intensive computing (DIC) provides a high performance computing approach to process large volume of data. In this study, a new formalization is introduced to present the two-stage DIC task execution in a stream manner. A novel heuristic algorithm is proposed for the scheduling problem due to the NP complexity. The theoretical approximation ratio bounds for the heuristic are analyzed and confirmed by the experimental evaluation. Overall, we observe that the proposed method conducts average 1.2 times makespan than the theoretic bound of the optimal solution. Besides, the proposed method outperforms the GA and FIFO scheduling schemes with overall improvements.

    AB - Data intensive computing (DIC) provides a high performance computing approach to process large volume of data. In this study, a new formalization is introduced to present the two-stage DIC task execution in a stream manner. A novel heuristic algorithm is proposed for the scheduling problem due to the NP complexity. The theoretical approximation ratio bounds for the heuristic are analyzed and confirmed by the experimental evaluation. Overall, we observe that the proposed method conducts average 1.2 times makespan than the theoretic bound of the optimal solution. Besides, the proposed method outperforms the GA and FIFO scheduling schemes with overall improvements.

    KW - Data intensive computing

    KW - Makespan

    KW - NP-hard

    KW - Scheduling

    UR - http://www.scopus.com/inward/record.url?scp=85012868497&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85012868497&partnerID=8YFLogxK

    U2 - 10.1016/j.jcss.2017.01.005

    DO - 10.1016/j.jcss.2017.01.005

    M3 - Article

    AN - SCOPUS:85012868497

    JO - Journal of Computer and System Sciences

    JF - Journal of Computer and System Sciences

    SN - 0022-0000

    ER -