Modeling and control design for performance management of web servers via an LPV approach

Wubi Qin, Qian Wang

    Research output: Contribution to journalArticle

    49 Citations (Scopus)

    Abstract

    This paper presents a control-theoretic approach to the performance management of Internet Web servers to meet service-level agreements. In particular, a CPU frequency management problem is studied to provide response time guarantees with minimal energy cost. It is argued that linear time-invariant modeling and control may not be sufficient for the system to adapt to dynamically varying load conditions. Instead, a Linear-parameter-varying (LPV) approach is presented in this paper, where workload arrival and service parameters are chosen as scheduling parameters to characterize time-varying operating conditions. Modeling the performance management of a Web server as an LPV system has been extensively discussed in this paper; we have derived first-principles models based on analyzing transient and steady-state queueing dynamics as well as empirical models using system identification algorithms. LPV H controllers are then designed for the derived LPV system models. Using real Web server workloads, the performance of LPV control compares favorably to various linear control designs and a design based on the conventional queueing theory. The proposed LPV modeling and control framework can be generalized to incorporate more sophisticated workload models and more complicated server environments. In addition, due to the LPV nature of Web systems with respect to load conditions, the proposed approach can be applied to a variety of resource management problems and used for middleware designs.

    Original languageEnglish (US)
    Pages (from-to)259-275
    Number of pages17
    JournalIEEE Transactions on Control Systems Technology
    Volume15
    Issue number2
    DOIs
    StatePublished - Mar 1 2007

    Fingerprint

    Servers
    Queueing theory
    Response time (computer systems)
    Middleware
    Program processors
    Dynamic models
    Identification (control systems)
    Scheduling
    Internet
    Controllers
    Costs

    All Science Journal Classification (ASJC) codes

    • Control and Systems Engineering
    • Electrical and Electronic Engineering

    Cite this

    @article{7f62c3aa0bc347328b2dfeeb15a9c1b4,
    title = "Modeling and control design for performance management of web servers via an LPV approach",
    abstract = "This paper presents a control-theoretic approach to the performance management of Internet Web servers to meet service-level agreements. In particular, a CPU frequency management problem is studied to provide response time guarantees with minimal energy cost. It is argued that linear time-invariant modeling and control may not be sufficient for the system to adapt to dynamically varying load conditions. Instead, a Linear-parameter-varying (LPV) approach is presented in this paper, where workload arrival and service parameters are chosen as scheduling parameters to characterize time-varying operating conditions. Modeling the performance management of a Web server as an LPV system has been extensively discussed in this paper; we have derived first-principles models based on analyzing transient and steady-state queueing dynamics as well as empirical models using system identification algorithms. LPV H∞ controllers are then designed for the derived LPV system models. Using real Web server workloads, the performance of LPV control compares favorably to various linear control designs and a design based on the conventional queueing theory. The proposed LPV modeling and control framework can be generalized to incorporate more sophisticated workload models and more complicated server environments. In addition, due to the LPV nature of Web systems with respect to load conditions, the proposed approach can be applied to a variety of resource management problems and used for middleware designs.",
    author = "Wubi Qin and Qian Wang",
    year = "2007",
    month = "3",
    day = "1",
    doi = "10.1109/TCST.2006.886433",
    language = "English (US)",
    volume = "15",
    pages = "259--275",
    journal = "IEEE Transactions on Control Systems Technology",
    issn = "1063-6536",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "2",

    }

    Modeling and control design for performance management of web servers via an LPV approach. / Qin, Wubi; Wang, Qian.

    In: IEEE Transactions on Control Systems Technology, Vol. 15, No. 2, 01.03.2007, p. 259-275.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Modeling and control design for performance management of web servers via an LPV approach

    AU - Qin, Wubi

    AU - Wang, Qian

    PY - 2007/3/1

    Y1 - 2007/3/1

    N2 - This paper presents a control-theoretic approach to the performance management of Internet Web servers to meet service-level agreements. In particular, a CPU frequency management problem is studied to provide response time guarantees with minimal energy cost. It is argued that linear time-invariant modeling and control may not be sufficient for the system to adapt to dynamically varying load conditions. Instead, a Linear-parameter-varying (LPV) approach is presented in this paper, where workload arrival and service parameters are chosen as scheduling parameters to characterize time-varying operating conditions. Modeling the performance management of a Web server as an LPV system has been extensively discussed in this paper; we have derived first-principles models based on analyzing transient and steady-state queueing dynamics as well as empirical models using system identification algorithms. LPV H∞ controllers are then designed for the derived LPV system models. Using real Web server workloads, the performance of LPV control compares favorably to various linear control designs and a design based on the conventional queueing theory. The proposed LPV modeling and control framework can be generalized to incorporate more sophisticated workload models and more complicated server environments. In addition, due to the LPV nature of Web systems with respect to load conditions, the proposed approach can be applied to a variety of resource management problems and used for middleware designs.

    AB - This paper presents a control-theoretic approach to the performance management of Internet Web servers to meet service-level agreements. In particular, a CPU frequency management problem is studied to provide response time guarantees with minimal energy cost. It is argued that linear time-invariant modeling and control may not be sufficient for the system to adapt to dynamically varying load conditions. Instead, a Linear-parameter-varying (LPV) approach is presented in this paper, where workload arrival and service parameters are chosen as scheduling parameters to characterize time-varying operating conditions. Modeling the performance management of a Web server as an LPV system has been extensively discussed in this paper; we have derived first-principles models based on analyzing transient and steady-state queueing dynamics as well as empirical models using system identification algorithms. LPV H∞ controllers are then designed for the derived LPV system models. Using real Web server workloads, the performance of LPV control compares favorably to various linear control designs and a design based on the conventional queueing theory. The proposed LPV modeling and control framework can be generalized to incorporate more sophisticated workload models and more complicated server environments. In addition, due to the LPV nature of Web systems with respect to load conditions, the proposed approach can be applied to a variety of resource management problems and used for middleware designs.

    UR - http://www.scopus.com/inward/record.url?scp=33947690698&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33947690698&partnerID=8YFLogxK

    U2 - 10.1109/TCST.2006.886433

    DO - 10.1109/TCST.2006.886433

    M3 - Article

    AN - SCOPUS:33947690698

    VL - 15

    SP - 259

    EP - 275

    JO - IEEE Transactions on Control Systems Technology

    JF - IEEE Transactions on Control Systems Technology

    SN - 1063-6536

    IS - 2

    ER -