An AnyLogic simulation model for power and performance analysis of data centres

Björn Frits Postema, Boudewijn R.H.M. Haverkort

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

    7 Citations (Scopus)
    926 Downloads (Pure)

    Abstract

    In this paper we propose a simulation framework that allows for the analysis of power and performance trade-offs for data centres that save energy via power management. The models are cooperating discrete-event and agent-based models, which enable a variety of data centre configurations, including various infrastructural choices, workload models, (heterogeneous) servers and power management strategies. The capabilities of our modelling and simulation approach is shown with an example of a 200-server cluster. A validation that compares our results, for a restricted model with a previously published numerical model is also provided.
    Original languageUndefined
    Title of host publicationProceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015
    EditorsMarta Beltrán, William Knottenbelt, Jeremy Bradley
    Place of PublicationSwitzerland
    PublisherSpringer
    Pages258-272
    Number of pages15
    ISBN (Print)978-3-319-23266-9
    DOIs
    Publication statusPublished - Aug 2015

    Publication series

    NameLecture Notes in Computer Science
    PublisherSpringer International Publishing
    Volume9272
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Keywords

    • EWI-26243
    • Steady state analysis
    • Performance analysis
    • Power Management
    • Power-performance trade-off
    • Data centres
    • METIS-314949
    • Transient analysis
    • Simulation
    • IR-98113
    • Cascading effect
    • Agent-based models
    • Discrete-event models

    Cite this

    Postema, B. F., & Haverkort, B. R. H. M. (2015). An AnyLogic simulation model for power and performance analysis of data centres. In M. Beltrán, W. Knottenbelt, & J. Bradley (Eds.), Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015 (pp. 258-272). (Lecture Notes in Computer Science; Vol. 9272). Switzerland: Springer. https://doi.org/10.1007/978-3-319-23267-6_17
    Postema, Björn Frits ; Haverkort, Boudewijn R.H.M. / An AnyLogic simulation model for power and performance analysis of data centres. Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015. editor / Marta Beltrán ; William Knottenbelt ; Jeremy Bradley. Switzerland : Springer, 2015. pp. 258-272 (Lecture Notes in Computer Science).
    @inproceedings{521a097996344f5a989f984480c7517f,
    title = "An AnyLogic simulation model for power and performance analysis of data centres",
    abstract = "In this paper we propose a simulation framework that allows for the analysis of power and performance trade-offs for data centres that save energy via power management. The models are cooperating discrete-event and agent-based models, which enable a variety of data centre configurations, including various infrastructural choices, workload models, (heterogeneous) servers and power management strategies. The capabilities of our modelling and simulation approach is shown with an example of a 200-server cluster. A validation that compares our results, for a restricted model with a previously published numerical model is also provided.",
    keywords = "EWI-26243, Steady state analysis, Performance analysis, Power Management, Power-performance trade-off, Data centres, METIS-314949, Transient analysis, Simulation, IR-98113, Cascading effect, Agent-based models, Discrete-event models",
    author = "Postema, {Bj{\"o}rn Frits} and Haverkort, {Boudewijn R.H.M.}",
    note = "eemcs-eprint-26243",
    year = "2015",
    month = "8",
    doi = "10.1007/978-3-319-23267-6_17",
    language = "Undefined",
    isbn = "978-3-319-23266-9",
    series = "Lecture Notes in Computer Science",
    publisher = "Springer",
    pages = "258--272",
    editor = "Marta Beltr{\'a}n and William Knottenbelt and Jeremy Bradley",
    booktitle = "Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015",

    }

    Postema, BF & Haverkort, BRHM 2015, An AnyLogic simulation model for power and performance analysis of data centres. in M Beltrán, W Knottenbelt & J Bradley (eds), Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015. Lecture Notes in Computer Science, vol. 9272, Springer, Switzerland, pp. 258-272. https://doi.org/10.1007/978-3-319-23267-6_17

    An AnyLogic simulation model for power and performance analysis of data centres. / Postema, Björn Frits; Haverkort, Boudewijn R.H.M.

    Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015. ed. / Marta Beltrán; William Knottenbelt; Jeremy Bradley. Switzerland : Springer, 2015. p. 258-272 (Lecture Notes in Computer Science; Vol. 9272).

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

    TY - GEN

    T1 - An AnyLogic simulation model for power and performance analysis of data centres

    AU - Postema, Björn Frits

    AU - Haverkort, Boudewijn R.H.M.

    N1 - eemcs-eprint-26243

    PY - 2015/8

    Y1 - 2015/8

    N2 - In this paper we propose a simulation framework that allows for the analysis of power and performance trade-offs for data centres that save energy via power management. The models are cooperating discrete-event and agent-based models, which enable a variety of data centre configurations, including various infrastructural choices, workload models, (heterogeneous) servers and power management strategies. The capabilities of our modelling and simulation approach is shown with an example of a 200-server cluster. A validation that compares our results, for a restricted model with a previously published numerical model is also provided.

    AB - In this paper we propose a simulation framework that allows for the analysis of power and performance trade-offs for data centres that save energy via power management. The models are cooperating discrete-event and agent-based models, which enable a variety of data centre configurations, including various infrastructural choices, workload models, (heterogeneous) servers and power management strategies. The capabilities of our modelling and simulation approach is shown with an example of a 200-server cluster. A validation that compares our results, for a restricted model with a previously published numerical model is also provided.

    KW - EWI-26243

    KW - Steady state analysis

    KW - Performance analysis

    KW - Power Management

    KW - Power-performance trade-off

    KW - Data centres

    KW - METIS-314949

    KW - Transient analysis

    KW - Simulation

    KW - IR-98113

    KW - Cascading effect

    KW - Agent-based models

    KW - Discrete-event models

    U2 - 10.1007/978-3-319-23267-6_17

    DO - 10.1007/978-3-319-23267-6_17

    M3 - Conference contribution

    SN - 978-3-319-23266-9

    T3 - Lecture Notes in Computer Science

    SP - 258

    EP - 272

    BT - Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015

    A2 - Beltrán, Marta

    A2 - Knottenbelt, William

    A2 - Bradley, Jeremy

    PB - Springer

    CY - Switzerland

    ER -

    Postema BF, Haverkort BRHM. An AnyLogic simulation model for power and performance analysis of data centres. In Beltrán M, Knottenbelt W, Bradley J, editors, Proceedings of the 12th European Workshop on Computer Performance Engineering, EPEW 2015. Switzerland: Springer. 2015. p. 258-272. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-23267-6_17