计算机应用 ›› 2010, Vol. 30 ›› Issue (07): 1742-1745.

• 网络与通信 • 上一篇    下一篇

基于预测机制的自适应负载均衡算法

石磊1,何增辉2   

  1. 1. 郑州大学南校区信息工程学院
    2. 郑州大学信息工程学院;河南省信息网络重点开放实验室
  • 收稿日期:2010-01-20 修回日期:2010-03-09 发布日期:2010-07-01 出版日期:2010-07-01
  • 通讯作者: 石磊
  • 基金资助:
    国家自然科学基金资助

Adaptive load balancing model based on prediction mechanism

,   

  • Received:2010-01-20 Revised:2010-03-09 Online:2010-07-01 Published:2010-07-01
  • Supported by:
    The National Natural Science Foundation

摘要: 工作负载特征对Web服务器集群中负载均衡调度算法的性能有重要影响。针对负载特征在调度算法所起作用的分析和讨论,提出基于预测机制的自适应负载均衡算法(RR_MMMCS-A-P)。通过监测工作负载,预测后续请求到达率和请求大小,快速调整相应参数,实现集群中各服务器之间的负载均衡。实验表明,无论是对计算密集型任务还是数据密集型任务,RR_MMMCS-A-P同基于CPU和CPU-MEM的调度算法相比在缩短平均响应时间方面具有较好的性能。

关键词: 集群, 工作负载, 负载均衡, 自适应, 预测机制

Abstract: Workload characteristics have an important impact on the performance of load balancing scheduling algorithms in Web server cluster systems. According to the analysis and discussion on the role of load characteristics for scheduling algorithm, a predictionbased adaptive load balancing algorithm (RR_MMMCS-A-P) was proposed in this paper. RR_MMMCS-A-P can predict the arrival rate and the size of the followup request by monitoring the workload characteristics and rapid adjustment of the corresponding parameters in order to balance the load between servers. The experimental results show that compared with CPUbased and CPUmemorybased scheduling algorithm, RR_MMMCS-A-P has better performance in reducing average response time for both calculationintensive and dataintensive jobs.

Key words: Clusters, Workload, Load Balance, Adaptive, Prediction Mechanism