摘要
本文总结了目前基于L7的Web群集负载平衡调度研究,分析了影响性能的主要因素。在估计Web负载时考虑了请求强度以及Web服务器自身的性能,提出了处理能力异构服务器群集的最小负载调度算法。在算法中还同时考虑了服务器在进入临界状态时性能急剧下降的因素,避免群集进入临界状态,新算法能较为准确地跟踪群集系统的负载,更好地均衡分配负载。
This paper summarizes the current research on the loadbalancing scheduling policy for Web clustering based on L7 of the OSI application layer, and analyses the main factors that affect performance.We consider the client requests on the basis of their expected impact on the server resources.Based on the analysis of server performance and client request types,we propose a new scheduling policy which takes heterogeneity into consideration. Through tracedriven simulation, the new scheduling policy can balance the load and achieve good Web clustering performance.
出处
《计算机工程与科学》
CSCD
2003年第3期7-10,共4页
Computer Engineering & Science