一种改进的流媒体集群动态负载均衡调度算法

摘要

In order to improve the timeliness of the node load feedback of streaming media cluster and the processing efficien-cy of numerous of requests in bursts,an improved algorithm based on dynamic feedback is proposed.In order to improve the effect of the cluster's load balancing,this method dynamically modifies the feedback period by the variation of tasks on the cluster nodes, and classifies them according to their load status.Experiments show that the proposed algorithm can effectively solve the problem of the cluster load tilt caused by numerous of concurrent requests,and improve the load balance effect of the streaming media cluster.%为提升流媒体集群节点负载反馈的及时性以及大量并发请求的处理效率问题,提出一种基于动态反馈的改进算法.该方法通过集群节点任务数变化量来动态修改反馈周期,同时根据其负载状况进行分类,以提高集群的负载均衡效果.实验表明,该算法能有效解决大量并发请求导致的集群负载倾斜问题,能够改善流媒体集群的负载均衡效果.