计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (15): 125-127.

• 网络、通信与安全 • 上一篇    下一篇

M/P/C/K排队系统呼损性能研究

符 涛,林 峰1,黄生叶1,叶 梧2,冯穗力   

  1. 1.湖南大学 计算机与通信学院,长沙 410082
    2.华南理工大学 电子与信息工程学院,广州 510640
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-05-21 发布日期:2007-05-21
  • 通讯作者: 符 涛

Call blocking performance study of M/P/C/K queuing system

FU Tao1,LIN Feng1,HUANG Sheng-ye1,YE Wu2,FENG Sui-li2   

  1. 1.College of Computer and Communication,Hunan University,Changsha 410082,China
    2.College of Electronics and Information Engineering,South China University of Technology,Guangzhou 510640,China
    E-mail:futaotao@126.com
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-21 Published:2007-05-21
  • Contact: 符 涛

摘要: 大量的研究表明网络中的业务量呈现自相似特征,但精确分析自相似业务流下的含多个服务装置、多缓存系统的排队性能非常困难。通过仿真得到服务时间是Pareto分布的M/P/C/K模型的平均呼损率(CBP),将它与服务时间是指数分布的传统网络排队模型M/M/C/K的CBP公式比较,结果表明,在有缓存情况下,自相似业务量导致网络的排队呼损性能劣化。

关键词: 自相似业务, Pareto分布, 指数分布, 呼损率

Abstract: Recent studies show that network traffic is self-similar.But it is difficult to analyse precisely the queuing performance of self-similar traffic in multi-channel and multi-buffer system.This paper derives the average Call Blocking Probability(CBP) of M/P/C/K model,and compares it with the CBP formula of M/M/C/K model.The experimental results in a system with buffers show that the self-similar traffic results in call blocking performance deterioration.

Key words: self-similar traffic, Pareto distribution, exponential distribution, Call Blocking Probability(CBP)