高级检索
    叶晓国 姜爱全 吴家皋. ANLMCC——一种基于主动网的分层多播拥塞控制方案[J]. 计算机研究与发展, 2005, 42(2): 273-279.
    引用本文: 叶晓国 姜爱全 吴家皋. ANLMCC——一种基于主动网的分层多播拥塞控制方案[J]. 计算机研究与发展, 2005, 42(2): 273-279.
    Ye Xiaoguo, Jiang Aiquan, and Wu Jiagao. ANLMCC—An Active Network-Based Layered Multicast Congestion Control Scheme[J]. Journal of Computer Research and Development, 2005, 42(2): 273-279.
    Citation: Ye Xiaoguo, Jiang Aiquan, and Wu Jiagao. ANLMCC—An Active Network-Based Layered Multicast Congestion Control Scheme[J]. Journal of Computer Research and Development, 2005, 42(2): 273-279.

    ANLMCC——一种基于主动网的分层多播拥塞控制方案

    ANLMCC—An Active Network-Based Layered Multicast Congestion Control Scheme

    • 摘要: 多媒体多播应用在Internet上的广泛部署对多播拥塞控制提出了要求.分层多播是适应网络异构性较有效的方案.针对现有分层多播大多存在拥塞响应延时大、吞吐率抖动剧烈和不满足TCP友好的问题,提出一种新的基于主动网的分层多播拥塞控制方案(ANLMCC),利用主动网灵活的服务定制能力,采用主动标记分层、优先级分层过滤,以及主动节点间逐跳的交互信令机制,大大改进了分层多播的性能.仿真实验表明,ANLMCC具有较快的拥塞响应速度、较好的稳定性和TCP友好的优点.

       

      Abstract: Effective multicast congestion control mechanism is urgently needed with the deployment of multimedia multicast application in Internet. Layered multicast is considered to be an efficient approach to cope with the network heterogeneity. Receiver-driven layered multicast schemes cannot respond to the congestion quickly because of both the delay for detecting congestion and the long IGMP leave delay. And they are significantly instable and unfriendly to TCP. In this paper, a novel congestion control scheme for layered multicast, called active network-based layered multicast congestion control scheme (ANLMCC), is proposed, taking advantage of the programmability and flexible service customization capability of active network. ANLMCC adopts novel mechanisms such as active layering with mark, filtering layer with priority, and hop-by-hop interactive signaling mechanism between the active nodes. Simulation results show that ANLMCC can respond to the congestion more quickly and is more stable and TCP-friendly, and can greatly improve the performance of layered multicast applications.

       

    /

    返回文章
    返回