• DocumentCode
    176171
  • Title

    Research of an adaptive PI admission control of Web service

  • Author

    Zhu Zijie ; Gao Xiang ; Zhang Chunhui

  • Author_Institution
    28th Res. Instn., China Electron. Technol. Group Corp., Nanjing, China
  • fYear
    2014
  • fDate
    May 31 2014-June 2 2014
  • Firstpage
    2290
  • Lastpage
    2293
  • Abstract
    A session-based admission time ratio (ATR) feedback control mechanism is introduced in order to the control Internet load pressure. An auto-regressive moving average (CARIMA) system is described as the Internet load pressure model. The Model Reference Adaptive Controller (MRAC) based on feedback control and feedforward compensation is designed to adjust ATR, and feedback controller is used to statue feedback control, also feedforward compensation is composed by error integration compensation. The simulation results of Apache Web server compared with MRAC and PI MRAC control methods show that the PI MRAC control based on system identification online has a better performance on improving the Internet quality of service (QoS) in case of overload.
  • Keywords
    PI control; Web services; autoregressive moving average processes; compensation; control system synthesis; feedback; feedforward; model reference adaptive control systems; pressure control; quality of service; ATR feedback control mechanism; Apache Web server; CARIMA system; Internet load pressure control; Internet load pressure model; MRAC design; QoS; Web service; adaptive PI admission control; conditional autoregressive moving average system; error integration compensation; feedforward compensation; model reference adaptive controller; proportional-integral control; quality of service; session-based admission time ratio; Adaptation models; Admission control; Feedback control; Internet; Load modeling; Mathematical model; ATR; CARIMA; PI MRAC control; admission control; system identify;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control and Decision Conference (2014 CCDC), The 26th Chinese
  • Conference_Location
    Changsha
  • Print_ISBN
    978-1-4799-3707-3
  • Type

    conf

  • DOI
    10.1109/CCDC.2014.6852552
  • Filename
    6852552