摘要
为了提高web集群负载均衡的效果，结合web服务用户访问静动态内容的特征，提出了一种基于访问特征负载预测的负载均衡算法。首先建立网络带宽负载和CPU、内存综合性能负载的小波包一支持向量机回归混合预测模型；然后根据用户请求的类型，结合负载预测的结果对任务进行分配和调度。仿真结果表明：与传统的基于负载预测的负载均衡算法相比，基于访问特征负载预测的负载均衡算法能达到更好的负载均衡效果，从而有效提高web集群的整体性能。
To obtain a better capacity of clustered web servers, a load balancing algorithm based on access characters-load prediction is presented in this paper by using the different characteristics of web server users＇ requests for static and dynamic contents. First, wavelet packet-SVR （Support Vector Regression） prediction model is constructed. Then, the arriving assignments are distributed according to the type of the requests and the load predicted results. The simulation results show that load balancing algorithm based on access characters-load prediction has better efficiency than traditional load balancing algorithm based on load prediction, meanwhile, it can improve the holistic performance of clustered web servers.
出处
《装甲兵工程学院学报》
2009年第5期68-71,共4页
Journal of Academy of Armored Force Engineering
