A performance analysis of EC2 cloud computing services for scientific computing

S. Ostermann, A. Iosup, M.N. Yigitbasi, R. Prodan, T. Fahringer, D.H.J. Epema

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

    303 Citations (Scopus)
    3 Downloads (Pure)

    Abstract

    Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.
    Original languageEnglish
    Title of host publicationCloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers)
    EditorsD. Avresky, M. Diaz, A. Bode, C. Bruno, E. Dekel
    Place of PublicationBerlin
    PublisherSpringer
    Pages115-131
    ISBN (Print)978-3-642-12635-2
    DOIs
    Publication statusPublished - 2010

    Publication series

    NameLecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering
    Volume34
    ISSN (Print)1867-8211

    Fingerprint

    Natural sciences computing
    Cloud computing
    Supercomputers
    Computer hardware
    Virtualization

    Cite this

    Ostermann, S., Iosup, A., Yigitbasi, M. N., Prodan, R., Fahringer, T., & Epema, D. H. J. (2010). A performance analysis of EC2 cloud computing services for scientific computing. In D. Avresky, M. Diaz, A. Bode, C. Bruno, & E. Dekel (Eds.), Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers) (pp. 115-131). (Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering; Vol. 34). Berlin: Springer. https://doi.org/10.1007/978-3-642-12636-9_9
    Ostermann, S. ; Iosup, A. ; Yigitbasi, M.N. ; Prodan, R. ; Fahringer, T. ; Epema, D.H.J. / A performance analysis of EC2 cloud computing services for scientific computing. Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers). editor / D. Avresky ; M. Diaz ; A. Bode ; C. Bruno ; E. Dekel. Berlin : Springer, 2010. pp. 115-131 (Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering).
    @inproceedings{1e1b1d8a0a4a4944919f1a8e4546a7ac,
    title = "A performance analysis of EC2 cloud computing services for scientific computing",
    abstract = "Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.",
    author = "S. Ostermann and A. Iosup and M.N. Yigitbasi and R. Prodan and T. Fahringer and D.H.J. Epema",
    year = "2010",
    doi = "10.1007/978-3-642-12636-9_9",
    language = "English",
    isbn = "978-3-642-12635-2",
    series = "Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering",
    publisher = "Springer",
    pages = "115--131",
    editor = "D. Avresky and M. Diaz and A. Bode and C. Bruno and E. Dekel",
    booktitle = "Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers)",
    address = "Germany",

    }

    Ostermann, S, Iosup, A, Yigitbasi, MN, Prodan, R, Fahringer, T & Epema, DHJ 2010, A performance analysis of EC2 cloud computing services for scientific computing. in D Avresky, M Diaz, A Bode, C Bruno & E Dekel (eds), Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers). Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering, vol. 34, Springer, Berlin, pp. 115-131. https://doi.org/10.1007/978-3-642-12636-9_9

    A performance analysis of EC2 cloud computing services for scientific computing. / Ostermann, S.; Iosup, A.; Yigitbasi, M.N.; Prodan, R.; Fahringer, T.; Epema, D.H.J.

    Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers). ed. / D. Avresky; M. Diaz; A. Bode; C. Bruno; E. Dekel. Berlin : Springer, 2010. p. 115-131 (Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering; Vol. 34).

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

    TY - GEN

    T1 - A performance analysis of EC2 cloud computing services for scientific computing

    AU - Ostermann, S.

    AU - Iosup, A.

    AU - Yigitbasi, M.N.

    AU - Prodan, R.

    AU - Fahringer, T.

    AU - Epema, D.H.J.

    PY - 2010

    Y1 - 2010

    N2 - Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.

    AB - Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.

    U2 - 10.1007/978-3-642-12636-9_9

    DO - 10.1007/978-3-642-12636-9_9

    M3 - Conference contribution

    SN - 978-3-642-12635-2

    T3 - Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering

    SP - 115

    EP - 131

    BT - Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers)

    A2 - Avresky, D.

    A2 - Diaz, M.

    A2 - Bode, A.

    A2 - Bruno, C.

    A2 - Dekel, E.

    PB - Springer

    CY - Berlin

    ER -

    Ostermann S, Iosup A, Yigitbasi MN, Prodan R, Fahringer T, Epema DHJ. A performance analysis of EC2 cloud computing services for scientific computing. In Avresky D, Diaz M, Bode A, Bruno C, Dekel E, editors, Cloud Computing (First International Conference, CloudComp 2009, Munich, Germany, October 19-21, 2009. Revised Selected Papers). Berlin: Springer. 2010. p. 115-131. (Lecture Notes of the Institute for Computer Sciences : Social Informatics and Telecommunications Engineering). https://doi.org/10.1007/978-3-642-12636-9_9