多种资源下结构化P2P网络的负载均衡方法

在真实网络环境中, 存在多种瓶颈资源, 如存储能力、处理请求能力和带宽等。传统的负载均衡方法只考虑了一种瓶颈资源, 忽视了其他瓶颈资源对网络性能的影响。针对该问题, 建立相似度概念模型, 设计用于虚拟服务器转移的转移代价函数, 提出可用于任意结构化P2P网络的多种资源负载均衡(MRLB)算法。仿真结果显示, 节点上任一资源的负载都随着节点能力的提升而相应增加。MRLB算法能够有效解决传统方法的缺点, 实现多种瓶颈资源的负载均衡。

Load balancing method in structured P2P network with multiple resources

There are multiple bottleneck resources in a real network environment, such as storage capacity, ability of proces-sing request and bandwidth. The traditional load balance methods only considered single resource, and ignored the influence from other bottleneck resources on network performance. To address this issue, this paper established similarity concept model, then developed multiple resources load balance (MRLB) algorithm for the arbitrary structured P2P network which used transferred cost function designed for virtual server transfer. Simulation results indicate that resource loads on a node can increase with the improvement of node capacity. So MRLB algorithm effectively solves the shortcomings of traditional methods and achieves load balance with multiple bottleneck resources.