自适应PI主动队列管理算法
作者:
基金项目:

Supported by the National Natural Science Foundation of China under Grant Nos.90104001,90204005(国家自然科学基金)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [13]
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    主动队列管理是一个非常活跃的研究领域,相对于丢尾算法,AQM(active queue management)能够提供更短的平均队列延迟和更高的带宽利用率.虽然PI(proporrional integral)主动队列管理算法的性能优于RED(random early detection)算法,但是PI算法的收敛速度比较慢.以PI算法为基础提出了一种自适应PI算法API(adaptive proportional integral).API通过实时测量链路的报文丢失率,获得当前的负载信息,然后动态设置PI算法中的有关参数.通过ns-2模拟表明,相对于PI及其改进算法PIP(proportional integral based series compensation and position feedback compensation),API具有更快的收敛速度和更小的队列抖动.

    Abstract:

    Active queue management (AQM) is a very active research area in networking. Compared with drop-tail, AQM can provide smaller average queue delay and higher bandwidth utilization. Although the performance of proportional integral (PI) controller is superior to that of random early detection (RED), its convergence speed is slow. This paper proposes an adaptive proportional integral (API) algorithm based on the original PI. API obtains load information by measuring the current packet-dropping rate, then sets PI parameters accordingly. Verified by using NS-2 simulations, API can achieve faster convergence speed and smaller queue oscillation than PI and PIP (proportional integral based series compensation and position feedback compensation) which is an improved algorithm of PI.

    参考文献
    [1]Braden B, Clark D, Crowcroft J, Davie B, Deering S, Estrin D, Floyd S, Jacobson V, Minshall G, Partridge C, Peterson L,Ramakrishnan K, Shenker S, Wroclawski J, Zhang L. Recommendations on queue management and congestion avoidance in the Internet. RFC2309, Internet Engineering Task Force, 1998.
    [2]Floyd S, Jacobson V. Random early detection gateways for congestion avoidance. IEEE/ACM Trans. on Networking, 1993,1(4):397-413.
    [3]Hollot CV, Misra V, Towsley D, Gong W. A control theoretic analysis of RED. In: Ammar M, ed. Proc. of the IEEE INFOCOM.Anchorage: IEEE Communications Society, 2001. 1510-1519.
    [4]Le L, Aikat J, Jeffay K, Smith FD. The effects of active queue management on Web performance. In: Proc. of the ACM SIGCOMM2003. Karlsruhe, 2003. 265-276. http://www.cs.unc.edu/~jeffay/papers/SIGCOMM-03.pdf
    [5]Floyd S, Gummadi R, Shenker S. Adaptive RED: An algorithm for increasing the robustness of RED's active queue management.2001. http://www.icir. org/~floyd
    [6]Kunniyur S, Srikant R. A time scale decomposition approach to adaptive ECN marking. In: Ammar M, ed. Proc. of the IEEE INFOCOM. Anchorage: IEEE Communications Society, 2001. 1330-1339.
    [7]Athuraliya S, Low S, Li VH, Yin QH. REM: Active queue management. IEEE Network, 2001,15(3):48-53.
    [8]Misra V, Gong WB, Towsley D. Fluid-Based analysis of a network of AQM routers supporting TCP flows with an application to RED. In: Proc. of the ACM SIGCOMM 2000. Stockholm, 2000. 151-160. http://gaia.cs.umass.edu/fluid/
    [9]Hollot CV, Misra V, Towsley D, Gong W. On designing improved controllers for AQM routers supporting TCP flows. In: Ammar M, ed. Proc. of the IEEE INFOCOM. Anchorage: IEEE Communications Society, 2001. 1726-1734.
    [10]Zhang HY, Liu BH, Dou WH. Design of a robust active queue management algorithm based on feedback compensation. In: Proc. of the ACM SIGCOMM 2003. Karlsruhe, 2003.277-285. http://portal.acm.org/citation.cfm?id=863987
    [11]Padhye J, Firoiu V, Towsley D, Kurose J. Modeling TCP throughput: A simple model and its empirical validation. In: Oran D, ed.Proc. of the ACM SIGCOMM. Vancouver: ACM Press, 1998. 303-314.
    [12]El-Gendy MA, Shin KG. Equation-Based packet marking for assured forwarding services. In: Proc. of the IEEE INFOCOM 2002.New York, 2002. 845-854. http://kabru.eecs.umich.edu/papers/publications/2002/mgendy_infocom02.pdf
    [13]Ns-2 Network simulato. 2001. http://www.isi.edu/nsnam/ns
    相似文献
    引证文献
引用本文

卢锡城,张明杰,朱培栋.自适应PI主动队列管理算法.软件学报,2005,16(5):903-910

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2003-11-17
  • 最后修改日期:2004-06-10
文章二维码
您是第19829757位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京市海淀区中关村南四街4号,邮政编码:100190
电话:010-62562563 传真:010-62562533 Email:jos@iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号