[1]张莉敏,陈晓纯.基于多路由配置的 IP 快速恢复流量均衡方法[J].计算机技术与发展,2019,29(06):90-94.[doi:10. 3969 / j. issn. 1673-629X. 2019. 06. 019]
　ZHANG Li-min,CHEN Xiao-chun.A Traffic Balancing Method for IP Fast Reroute Based on Multiple Routing Configuration[J].,2019,29(06):90-94.[doi:10. 3969 / j. issn. 1673-629X. 2019. 06. 019]
点击复制

基于多路由配置的 IP 快速恢复流量均衡方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年06期
页码:
90-94
栏目:
智能、算法、系统工程
出版日期:
2019-06-10

文章信息/Info

Title:
A Traffic Balancing Method for IP Fast Reroute Based on Multiple Routing Configuration
文章编号:
1673-629X(2019)06-0090-05
作者:
张莉敏; 陈晓纯
广东理工学院 信息工程系,广东 肇庆 526100
Author(s):
ZHANG Li-min; CHEN Xiao-chun
Department of Information Engineering,Guangdong Polytechnic College,Zhaoqing 526100,China
关键词:
IP 快速恢复; 备份配置; 链路负载; 重路由; 关键节点
Keywords:
IP fast recovery; backup configuration; link load; reroute; Key Nodes
分类号:
TP393.02
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 06. 019
摘要:
针对 IP 网络故障后流量不均衡的问题,提出了一种新的备份配置设计方法来解决 IP 快速恢复重路由中的网络拥塞问题。 该方法利用网络拓扑结构和流量矩阵来创建备份配置以将高负载链路上的流量分离到其他链路。 文中设计了一种确定关键节点的方法,其主要思想是根据备份配置中节点的介数中心性和紧密性的取值来定义关键节点。 并且在选取关键节点时,考虑节点在网络中是否相邻,根据计算得到的值采用两种方法选取前 K 个节点当作关键节点,从而最大化关键节点的可用链路。 实验结果表明,与传统算法相比,该算法最大链路负载减少量约为 73%,单链路故障时最大减少跳数约为 45%。 另外,从仿真结果得出这样的结论:考虑大型网络的关键节点的位置并使用紧密性算法来选择关键节点是更好的策略。
Abstract:
Aiming at the unbalanced traffic after IP network failure,we present a new backup configuration design method to solve the problem of network congestion in IP fast recovery rerouting. This method creates the backup configuration to separate traffic on high load links to the other links by utilizing the network topology and traffic matrix. In this paper,a new method to determine Key Nodes is designed. The main idea is to define the Key Nodes according to the values of the betweenness centrality and closeness of the nodes in the backup configuration. When choosing the key nodes,considering whether the nodes are adjacent or not,two methods are used to select the first K nodes as the Key Nodes according to the values calculated previously,then maximizing available links of Key Nodes. The experiment shows that maximum link load reduction is approximately 73% compared with the state of conventional algorithm,and maximum reduced hops is about 45% when single link failure. In addition,we can conclude that considering the location of the Key Nodes for a large network is a better strategy when using the closeness algorithm to select the Key Nodes.
更新日期/Last Update: 2019-06-10