Rhizoma

A runtime for self-deploying, self-managing overlays

Qin Yin, Adrian Schüpbach, Justin Cappos, Andrew Baumann, Timothy Roscoe

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime ("Rhizoma") with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging "proving ground" for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.

    Original languageEnglish (US)
    Title of host publicationMiddleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings
    Pages184-204
    Number of pages21
    Volume5896 LNCS
    DOIs
    StatePublished - 2009
    EventACM/IFIP/USENIX 10th International Middleware Conference - Urbana, IL, United States
    Duration: Nov 30 2009Dec 4 2009

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume5896 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    OtherACM/IFIP/USENIX 10th International Middleware Conference
    CountryUnited States
    CityUrbana, IL
    Period11/30/0912/4/09

    Fingerprint

    Overlay
    Resources
    Availability
    Constrained optimization
    Electric fuses
    P2P Systems
    Requirements
    Virtual Machine
    Constrained Optimization Problem
    Resource Management
    Response Time
    Decentralized
    Pricing
    Express
    Infrastructure
    Maximise
    Optimization Problem
    Real-time
    Computing
    Costs

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Yin, Q., Schüpbach, A., Cappos, J., Baumann, A., & Roscoe, T. (2009). Rhizoma: A runtime for self-deploying, self-managing overlays. In Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings (Vol. 5896 LNCS, pp. 184-204). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5896 LNCS). https://doi.org/10.1007/978-3-642-10445-9_10

    Rhizoma : A runtime for self-deploying, self-managing overlays. / Yin, Qin; Schüpbach, Adrian; Cappos, Justin; Baumann, Andrew; Roscoe, Timothy.

    Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings. Vol. 5896 LNCS 2009. p. 184-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5896 LNCS).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Yin, Q, Schüpbach, A, Cappos, J, Baumann, A & Roscoe, T 2009, Rhizoma: A runtime for self-deploying, self-managing overlays. in Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings. vol. 5896 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5896 LNCS, pp. 184-204, ACM/IFIP/USENIX 10th International Middleware Conference, Urbana, IL, United States, 11/30/09. https://doi.org/10.1007/978-3-642-10445-9_10
    Yin Q, Schüpbach A, Cappos J, Baumann A, Roscoe T. Rhizoma: A runtime for self-deploying, self-managing overlays. In Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings. Vol. 5896 LNCS. 2009. p. 184-204. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-10445-9_10
    Yin, Qin ; Schüpbach, Adrian ; Cappos, Justin ; Baumann, Andrew ; Roscoe, Timothy. / Rhizoma : A runtime for self-deploying, self-managing overlays. Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings. Vol. 5896 LNCS 2009. pp. 184-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{d20b5c3a703d46fe86e4ab71d491bde5,
    title = "Rhizoma: A runtime for self-deploying, self-managing overlays",
    abstract = "The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime ({"}Rhizoma{"}) with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging {"}proving ground{"} for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.",
    author = "Qin Yin and Adrian Sch{\"u}pbach and Justin Cappos and Andrew Baumann and Timothy Roscoe",
    year = "2009",
    doi = "10.1007/978-3-642-10445-9_10",
    language = "English (US)",
    isbn = "3642104444",
    volume = "5896 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "184--204",
    booktitle = "Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings",

    }

    TY - GEN

    T1 - Rhizoma

    T2 - A runtime for self-deploying, self-managing overlays

    AU - Yin, Qin

    AU - Schüpbach, Adrian

    AU - Cappos, Justin

    AU - Baumann, Andrew

    AU - Roscoe, Timothy

    PY - 2009

    Y1 - 2009

    N2 - The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime ("Rhizoma") with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging "proving ground" for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.

    AB - The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime ("Rhizoma") with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging "proving ground" for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.

    UR - http://www.scopus.com/inward/record.url?scp=70549109057&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=70549109057&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-10445-9_10

    DO - 10.1007/978-3-642-10445-9_10

    M3 - Conference contribution

    SN - 3642104444

    SN - 9783642104442

    VL - 5896 LNCS

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 184

    EP - 204

    BT - Middleware 2009: ACM/IFIP/USENIX 10th International Middleware Conference, Proceedings

    ER -