• DocumentCode
    2305232
  • Title

    Research on E-SG mixed polling model based on two kinds of businesses

  • Author

    Jin Lu ; Hongwei Ding ; Dongming Zhou ; Yifan Zhao

  • Author_Institution
    Sch. of Inf., Yunnan Univ., Kunming, China
  • fYear
    2012
  • fDate
    29-31 Dec. 2012
  • Firstpage
    2019
  • Lastpage
    2023
  • Abstract
    Based on the E-G level two polling system [8], proposed E-SG (Exhaustive - Simplified Gated) polling system model controlled by two kinds of service strategies commonly, which were applied to two different priorities access. The model was abstracted as Markov process and the system state variables were defined as probability density function with N dimension. Then according to the Queuing service mechanism under stable condition, established the state transfer equation of probability generating function, and exactly analyzed the key characteristics of average queue length and average waiting delay, experimental conclusions compared with reference [8] illustrated the new mixed service strategy performed better fairness and stability, and satisfied the needs of network QoS better. The new model not only accomplished the effective control of priorities but also improved the performance of the system.
  • Keywords
    Markov processes; commerce; probability; quality of service; queueing theory; E-G level two polling system; E-SG mixed polling model; Markov process; average queue length; average waiting delay; businesses; exhaustive-simplified gated polling system model; experimental conclusions; mixed service strategy; network QoS; probability density function; probability generating function; queuing service mechanism; state transfer equation; system state variables; E-SG; QoS; polling system; queuing theory;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Network Technology (ICCSNT), 2012 2nd International Conference on
  • Conference_Location
    Changchun
  • Print_ISBN
    978-1-4673-2963-7
  • Type

    conf

  • DOI
    10.1109/ICCSNT.2012.6526315
  • Filename
    6526315