###
DOI:
计算机系统应用英文版:2011,20(5):47-50,74
本文二维码信息
码上扫一扫!
基于内容分类的集群负载均衡算法
(浙江科技学院 信息与电子工程学院,杭州 310023)
Content-Classification Load Balancing Algorithm in Cluster
(School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1670次   下载 2808
Received:August 20, 2010    Revised:October 17, 2010
中文摘要: 集群技术为Web 服务带来了新的解决方案,其核心思想是负载均衡策略。在分析已有方法的基础上,提出了一种基于内容分类的集群负载均衡算法。该算法通过对用户请求分类后均匀地分配给各节点服务器,以使每个节点服务器得到的各类请求量大致相同。同时算法引入了一个反馈环节,通过负载权值的等效变换和进入临界状态后的动态权值调整来防止节点负载倾斜。实验证明该算法能有效提高集群系统的整体性能。
中文关键词: 集群  负载均衡  内容分类  动态反馈  权值
Abstract:The cluster technology brings a new solution for the WEB service, the key point is load balancing strategy. Based on the existing algorithms, a content-classification load balancing algorithm is proposed. By using content classification, user requests could evenly distribute to the nodes in the cluster, so that each node would get roughly the same amount of various types of requests. To prevent the load skew, a feedback mechanism was introduced, furthermore, the weight of each node interval was adjusted by using an equivalent load-alternant and would multiplicative decrease in critical area. The test shows this algorithm could improve the performance of the cluster system.
文章编号:     中图分类号:    文献标志码:
基金项目:
引用文本:
郑祺,周广平.基于内容分类的集群负载均衡算法.计算机系统应用,2011,20(5):47-50,74
ZHENG Qi,ZHOU Guang-Ping.Content-Classification Load Balancing Algorithm in Cluster.COMPUTER SYSTEMS APPLICATIONS,2011,20(5):47-50,74