摘要
对IP多媒体子系统(IMS)中的实时计费系统(OCS)集群进行了研究,针对其高集群异构性和高请求负载异构性特征,提出了一种最小公倍数能力(LCMC)负载均衡算法。此算法考虑了各个服务器处理能力的差异和各个服务器当前的负载,并特别考虑了不同请求产生的负载差异,采用概率的方式进行请求分发。实验结果表明,在高集群异构性和高请求负载异构性的系统中,LCMC算法明显优于一些传统的负载均衡算法,能很好地满足IMS中OCS集群独有的特征。
The paper studies the online charging system （OCS） cluster in IP multimedia subsystems （IMS）, and in consideration of the high heterogeneity features of the OCS cluster, proposes a least common multiple capacity （LCMC） load-balancing algorithm. This algorithm considers the differences between processing capacity of each server, the run-time load of each server, and especially the differences between various request loads, distributing requests by the probability approach. The experimental results show that the LCMC algorithm significantly outperforms some traditional load-balancing algorithms in the system, of high heterogeneous cluster and high heterogeneous request load, and meets the unique characteristics of the OCS cluster in IMS.
出处
《高技术通讯》
EI
CAS
CSCD
北大核心
2009年第11期1117-1123,共7页
Chinese High Technology Letters
基金
973计划(2007CB307100
2007CB307103)
国家自然科学基金(60902051)
国家杰出青年科学基金(60525110)
电子信息产业发展基金资助项目
