Self-adaptive SLA-driven capacity management for internet services

Bruno Abrahao, Virgilio Almeida, Jussara Almeida, Zhang Alex, Dirk Beyer, Fereydoon Safai

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

    Abstract

    This work considers the problem of hosting multiple third-party Internet services in a cost-effective manner so as to maximize a provider's business objective. For this purpose, we present a dynamic capacity management framework based on an optimization model, which links a cost model based on SLA contracts with an analytical queuing-based performance model, in an attempt to adapt the platform to changing capacity needs in real time. In addition, we propose a two-level SLA specification for different operation modes, namely, normal and surge, which allows for per-use service accounting with respect to requirements of throughput and tail distribution response time. The cost model proposed is based on penalties, incurred by the provider due to SLA violation, and rewards, received when the service level expectations are exceeded. Finally, we evaluate approximations for predicting the performance of the hosted services under two different scheduling disciplines, namely FCFS and processor sharing. Through simulation, we assess the effectiveness of the proposed approach as well as the level of accuracy resulting from the performance model approximations.

    Original languageEnglish (US)
    Title of host publication10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006
    Pages557-568
    Number of pages12
    StatePublished - Dec 1 2006
    Event10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006 - Vancouver, BC, Canada
    Duration: Apr 3 2006Apr 7 2006

    Publication series

    NameIEEE Symposium Record on Network Operations and Management Symposium

    Conference

    Conference10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006
    CountryCanada
    CityVancouver, BC
    Period4/3/064/7/06

    Fingerprint

    Internet
    Costs
    Scheduling
    Throughput
    Specifications
    Industry

    ASJC Scopus subject areas

    • Hardware and Architecture
    • Electrical and Electronic Engineering

    Cite this

    Abrahao, B., Almeida, V., Almeida, J., Alex, Z., Beyer, D., & Safai, F. (2006). Self-adaptive SLA-driven capacity management for internet services. In 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006 (pp. 557-568). [1687584] (IEEE Symposium Record on Network Operations and Management Symposium).

    Self-adaptive SLA-driven capacity management for internet services. / Abrahao, Bruno; Almeida, Virgilio; Almeida, Jussara; Alex, Zhang; Beyer, Dirk; Safai, Fereydoon.

    10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006. 2006. p. 557-568 1687584 (IEEE Symposium Record on Network Operations and Management Symposium).

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

    Abrahao, B, Almeida, V, Almeida, J, Alex, Z, Beyer, D & Safai, F 2006, Self-adaptive SLA-driven capacity management for internet services. in 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006., 1687584, IEEE Symposium Record on Network Operations and Management Symposium, pp. 557-568, 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006, Vancouver, BC, Canada, 4/3/06.
    Abrahao B, Almeida V, Almeida J, Alex Z, Beyer D, Safai F. Self-adaptive SLA-driven capacity management for internet services. In 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006. 2006. p. 557-568. 1687584. (IEEE Symposium Record on Network Operations and Management Symposium).
    Abrahao, Bruno ; Almeida, Virgilio ; Almeida, Jussara ; Alex, Zhang ; Beyer, Dirk ; Safai, Fereydoon. / Self-adaptive SLA-driven capacity management for internet services. 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006. 2006. pp. 557-568 (IEEE Symposium Record on Network Operations and Management Symposium).
    @inproceedings{819e08a4f4c242859d2b0d2b81e174ad,
    title = "Self-adaptive SLA-driven capacity management for internet services",
    abstract = "This work considers the problem of hosting multiple third-party Internet services in a cost-effective manner so as to maximize a provider's business objective. For this purpose, we present a dynamic capacity management framework based on an optimization model, which links a cost model based on SLA contracts with an analytical queuing-based performance model, in an attempt to adapt the platform to changing capacity needs in real time. In addition, we propose a two-level SLA specification for different operation modes, namely, normal and surge, which allows for per-use service accounting with respect to requirements of throughput and tail distribution response time. The cost model proposed is based on penalties, incurred by the provider due to SLA violation, and rewards, received when the service level expectations are exceeded. Finally, we evaluate approximations for predicting the performance of the hosted services under two different scheduling disciplines, namely FCFS and processor sharing. Through simulation, we assess the effectiveness of the proposed approach as well as the level of accuracy resulting from the performance model approximations.",
    author = "Bruno Abrahao and Virgilio Almeida and Jussara Almeida and Zhang Alex and Dirk Beyer and Fereydoon Safai",
    year = "2006",
    month = "12",
    day = "1",
    language = "English (US)",
    isbn = "1424401429",
    series = "IEEE Symposium Record on Network Operations and Management Symposium",
    pages = "557--568",
    booktitle = "10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006",

    }

    TY - GEN

    T1 - Self-adaptive SLA-driven capacity management for internet services

    AU - Abrahao, Bruno

    AU - Almeida, Virgilio

    AU - Almeida, Jussara

    AU - Alex, Zhang

    AU - Beyer, Dirk

    AU - Safai, Fereydoon

    PY - 2006/12/1

    Y1 - 2006/12/1

    N2 - This work considers the problem of hosting multiple third-party Internet services in a cost-effective manner so as to maximize a provider's business objective. For this purpose, we present a dynamic capacity management framework based on an optimization model, which links a cost model based on SLA contracts with an analytical queuing-based performance model, in an attempt to adapt the platform to changing capacity needs in real time. In addition, we propose a two-level SLA specification for different operation modes, namely, normal and surge, which allows for per-use service accounting with respect to requirements of throughput and tail distribution response time. The cost model proposed is based on penalties, incurred by the provider due to SLA violation, and rewards, received when the service level expectations are exceeded. Finally, we evaluate approximations for predicting the performance of the hosted services under two different scheduling disciplines, namely FCFS and processor sharing. Through simulation, we assess the effectiveness of the proposed approach as well as the level of accuracy resulting from the performance model approximations.

    AB - This work considers the problem of hosting multiple third-party Internet services in a cost-effective manner so as to maximize a provider's business objective. For this purpose, we present a dynamic capacity management framework based on an optimization model, which links a cost model based on SLA contracts with an analytical queuing-based performance model, in an attempt to adapt the platform to changing capacity needs in real time. In addition, we propose a two-level SLA specification for different operation modes, namely, normal and surge, which allows for per-use service accounting with respect to requirements of throughput and tail distribution response time. The cost model proposed is based on penalties, incurred by the provider due to SLA violation, and rewards, received when the service level expectations are exceeded. Finally, we evaluate approximations for predicting the performance of the hosted services under two different scheduling disciplines, namely FCFS and processor sharing. Through simulation, we assess the effectiveness of the proposed approach as well as the level of accuracy resulting from the performance model approximations.

    UR - http://www.scopus.com/inward/record.url?scp=34250778250&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=34250778250&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:34250778250

    SN - 1424401429

    SN - 9781424401420

    T3 - IEEE Symposium Record on Network Operations and Management Symposium

    SP - 557

    EP - 568

    BT - 10th IEEE/IFIP Network Operations and Management Symposium, NOMS 2006

    ER -