云计算环境下基于并行计算熵的负载均衡算法

Load Balance Algorism for Cloud Computing Environment Based on Concurrency Computing Shannon

摘要:

针对云计算环境下大量并行任务运行所导致的某些节点负载过重,从而引起整个系统负载不均和效率低下的问题,提出了一种基于并行计算熵的资源负载均衡算法；首先,描述了云计算虚拟机部署原理并给出了适合云计算环境和异构集群的并行计算熵的计算方式,然后,定义了在系统并行计算熵低于阈值时迁移的源物理节点、迁移虚拟机和迁移目标物理节点的确定方式；最后,定义了基于并行计算熵的负载均衡算法；采用CloudSim云计算仿真工具对文中方法进行仿真实验,结果表明文中方法较其它方法的平均负载均衡度约低21.8%,具有较低的任务平均响应时间、合理的资源利用率和较小的负载均衡度,具有较大的优越性。

Abstract:

In order to solve the problems of lots of operating concurrent tasks leading to the problems such as heavy load balance for some nodes, and therefore leading to the unbalance load and low efficiency of the whole system, a algorithm based on concurrency computing Shannon was proposed. Firstly, the deploy principle for virtual machine in Cloud computing environment was given and the computing for concurrency computing Shannon suiting for cloud computing and heteroid machine was given, then the migrating source physical node, migrating virtual machine and goal physical node when the system concurrency computing Shannon were introduced. Finally, the load balance algorism based on concurrency computing Shannon was defined. Using the cloud computing tool CloudSim to simulate the method in this paper, and it is proved it can effectively realize the node load balance, and compared with the other methods , it has the advantages such as low average response time, round resource usage and less load balance degree, so it has big priority. 