分布式计算机系统负载平衡研究

Study on Load Balancing of Distributed Computer System

负载平衡是分布式系统中的一个研究热点。为了实现和充分利用这种能力,需要优良的负载平衡分配方案。对负载问题进行了数学化描述,研究了分布计算机系统中基本负载平衡策略,在此基础上提出了一个基于人工智能的负载平衡方案。利用在线跟踪技术,获得作业的行为特征(资源需求和执行时间等),从而筛选出那些不值得转移的短作业。性能测试的结果表明,文中所提出的方法能够较好地缩短作业的平均响应时间和提高系统的资源利用率,实现了动态负载平衡的目的。

Load balancing is a hot reacarch point in distributed systems,but the good load balancing strategies is needed. Based on mathematics model description, the basic load balancing strategies and their characters are studied. In this groundworks,an intelligence load balancing strategy and its algorithm are addressed. Using on- line tracing technique, this paper can predicate the behavior of a job, such as its resource requirements and its approximate execution time. Therefore, we can recognize those short - lived jobs which are not worth transferring. Experiment measurement results show that it is able to reduce mean response time and improve resource utilization of systems.