面向ICN组播的基于感染力的汇聚点部署方法  

Rendezvous Point Deployment Method Based on Force of Infection For ICN Multicast

作　　者：田震宇 尤佳莉[1,2] 朱小勇 TIAN Zhenyu;YOU Jiali;ZHU Xiaoyong(National Network New Media Engineering Research Center,Institute of Acoustic,Chinese Academy of Science,Beijing,100190,China;University of Chinese Academy of Science,Beijing,100049,China)

机构地区：[1]中国科学院声学研究所国家网络新媒体工程技术研究中心,北京100190 [2]中国科学院大学,北京100049

出　　处：《网络新媒体技术》2024年第1期21-28,共8页Network New Media Technology

基　　金：面向信息中心网络的传输技术研究(编号:E155180201)。

摘　　要：在组播系统中,汇聚点(RP)作为组播树的根节点,负责将源内容引入组播树。RP汇集了网络中的流量,容易成为网络性能瓶颈。为了提高网络容纳的组播组数量(组播容量),可以将更多的网络节点升级为RP。相比于传统的IP组播,信息中心网络(ICN)组播大幅降低了RP管理代价,从而允许部署大量RP。因此,本文基于ICN组播展开研究。但是,由于设备升级成本和系统稳定性等原因,ICN组播中RP数量仍然不能太多。现有基于P中值模型法的RP部署方法没有充分考虑到RP之间的负载压力,容易导致RP计算资源浪费和一些RP的负载过重,限制了组播容量。因此,本文提出了一种面向ICN组播的基于感染力的RP部署方法。该方法首先提出了一种组播感染力指标,用于刻画RP之间的负载压力。其次,利用组播感染力,提出了一种基于贪心策略的RP部署方法,优先将负载压力小的节点升级为RP。2组实验证明,在部署15个RP的情况下,本方法相比P中值模型法,将组播容量提高3.48倍;在组播容量都达到9000组时,本方法所需的RP数量仅为p中值模型法的29.4%。In a multicast system,the Rendezvous Point(RP)serves as the root node of the multicast tree and is responsible for introducing source content into the multicast tree.RPs collect traffic in the network and can easily become a bottleneck for network performance.To increase the number of multicast groups accommodated by the network(multicast capacity),more network nodes can be upgraded to serve as RPs.Compared to traditional IP multicast,ICN multicast significantly reduces the cost of managing RPs,allowing for the deployment of a large number of RPs.Therefore,this paper focuses on ICN multicast research.However,due to device upgrade costs and system stability,the number of RPs in ICN multicast still cannot be too high.Existing RP deployment methods based on the P-median model do not fully consider the load pressure between RPs,which can lead to wasted RP computing resources and overloaded RPs,limiting multicast capacity.Therefore,this paper proposes an infection-based RP deployment method for ICN multicast.Firstly,a multicast infection force indicator is proposed to characterize the load pressure between RPs.Secondly,based on multicast infection force,a greedy strategy-based RP deployment method is proposed.This method prioritizes upgrading nodes with lower load pressure to serve as RPs.Two sets of experiments have shown that,with the deployment of 15 RPs,this method can increase multicast capacity by 3. 48 times compared to the P - median model, and when the multicast capacity reaches 9 000 groups, the number of RPs required by this method is only about 29. 4% of the P - median model.

关 键 词：组播 汇聚点 部署 组播容量 感染力 

分 类 号：TN929.5[电子电信—通信与信息系统]