LRV负载均衡算法的研究与实现

在研究基于软交换的集群媒体服务器的系统特征的基础上，提出了1种LRV(limited resource vector)负载均衡算法并提供了算法的软件实现方法，该算法考虑了异构系统中资源的协调使用并有效防止了系统性能的剧烈变化. 通过在负载均衡系统Petri网模型上的大量仿真模拟，对LRV负载均衡算法与其他传统负载均衡算法在系统吞吐量、系统响应时间和性能平稳性几个方面做了比较和分析. 结果表明，LRV算法具有更优越的性能，LRV算法的软件实现方法对其他异构集群系统也具有重要参考价值.

Research and Implementation of A Limited Resource Vector Load-Balancing Algorithm

Based on the system feature of softswitch-based heterogeneous clustered media server, this paper proposed a limited resource vector load-balancing algorithm and also provided its software implementation method. Harmonious usage of system resources was considered and violent shaking of the system performance was avoided in the algorithm. A lot of simulations on the Petri net model of balance system had been conducted and the algorithm had been compared with some traditional algorithms on system throughput, request response time and performance stability. Results of simulation experiments show that the algorithm has higher performance. The implementation method of the algorithm is also useful for other heterogeneous clustered systems.