摘要
MPLSTE的关键问题是LSP的分布优化问题,针对其NP-hard特性,提出了一种基于遗传算法的求解方法,同时配置了一组LSP。仿真结果表明,本算法在一定程度上改善了网络资源的利用情况,避免了网络拥塞,达到了负载均衡的目的。与此同时,本算法还克服了一条一条配置LSP产生的"顶端优势"问题,能以任意顺序建立LSP。
The key problem of the MPLS TE is the optimized distribution of LSPs. This paper provides a solution based on the genetic algorithm according to the NP - hard feature and configures a group of LSPs. The stimulation results prove that this algorithm effectively improves the usage of the network resources and avoids network congestion, thus leading to the load balance. In addition, this algorithm solves the problem of ＂top advantage＂ caused by the configuration of LSPs that is performed one by one. It can create .LSPs at a random order.
出处
《微计算机应用》
2007年第12期1242-1245,共4页
Microcomputer Applications
