图像切割法在缓存服务器中部署策略

Graph Partitioning Method to Determine Servers Placement

为了解决内容分发网络中因缓存服务器部署位置而导致的网络延时等问题,同时改进内容分发网络中缓存服务器的合理化部署策略,论文借鉴网络仿真图像轻点匹配算法的思想,提出了一种CDN图像切割的多层K划分模式,通过理论分析和计算,同时找到了一种适合CDN网络环境的最大化边缘切割的K路KL(Kernighan-Lin)细化算法,它是根据物理部署点抽象出网络节点拓扑图,再通过相应的算法进行粗化和细化阶段的处理,使节点部署达到最佳化,最后通过算法分析证明了在解决CDN缓存服务器部署问题上,多层K划分算法是一种可靠高效的方法。

In order to solve the problem of network latency caused by cache servers’ unreasonable placement in CDN （Content Delivery Network） ,by use of graph partitioning method to make search on CDN cache servers’ placement ,an idea that using graph partitioning to solve the problem is proposed through theoretical analysis and the specific algorithm of parti‐tioning is researched .In this paper ,by reference to light vertex matching algorithm of graph partitioning for network simula‐tion ,a multilevel K‐way algorithm of graph partitioning for CDN is proposed .The maximized edge cut K‐way KL refinement algorithm is discussed .Multilevel k‐way algorithm is a feasible algorithm for CDN graph partitioning .