一种新的结构化对等网负载均衡算法

摘要

研究结构化对等网(P2P)中的负载均衡问题,P2P网络的节点、延迟和处理能力差异性很大,当前负载均衡算法忽略节点差异性,造成网络负载极不平衡,容易出现“热点”问题,使负载变化大.为了更好的均衡P2P负载,提出一种新的网络负载均衡算法.算法充分考虑节点之间的差异性,对物理节点的地址空间进行动态分配,对于热点资源下载采用局部搜索算法找到邻居节点,并自动把负载转移到轻载节点上,保证节点间负载均衡.仿真结果表明,新网络负载均衡算法加快了负载均衡速度,使P2P网络负载均衡更加均衡,能够很好保持系统稳定性.%In order to better balance the P2P load, this paper proposed a new network load balancing algorithm. The algorithm considered the differences between the nodes, and dynamicly allocated the address space of the nodes. When hot resources were downloaded, local search algorithm was used to find the neighbor nodes, and automatically transferred die load to the light loaded nodes, thus ensuring the load balancing among nodes. The simulation results show that the network load balancing algorithm can speed up the load balancing speed, reduce the load balancing o-verhead, and well maintain the stability of the system.