| 基于mean-variance的服务集群负载均衡方法 |
| |
| 引用本文： | 包晓安,魏雪,陈磊,胡国亨,张娜.基于mean-variance的服务集群负载均衡方法[J].电信科学,2017(1):1-8. |
| |
| 作者姓名： | 包晓安 魏雪 陈磊 胡国亨 张娜 |
| |
| 基金项目： | 国家自然科学基金资助项目(61379036;61502430)，国家自然科学基金委中丹合作项目(61361136002)，浙江省重大科技专项重点工业项目(2014C01047)，浙江理工大学“521人才培养计划”基金资助项目 The National Natural Science Foundation of China(61379036;61502430)，China-Denmark Cooperation Program of the National Natural Science Foundation of China(61361136002)，Major Science and Technology Projects of Zhejiang Province(2014C01047)，521 Talent Project of Zhejiang Sci-Tech University |
| |
| 摘 要： | 大量并发请求任务进行分配时,负载调度机制是通过最小化响应时间及最大化节点利用率实现网络中节点的负载均衡,在基于遗传算法的负载均衡算法中,适应度函数设计对服务集群负载均衡效率产生重要的影响.对此提出了一种基于mean-variance的服务集群负载均衡方法对适应度函数进行优化,采用投资组合选择模型mean-variance进行最小化响应时间,以得到每个服务器资源利用率的权重,从而获得最优的分配组合,进而提高适应度函数的准确性和有效性.在不同服务环境下与其他模型进行比较,仿真结果表明,本文的负载均衡算法在节点利用率和响应时间方面使服务集群得到了更好的均衡.
|
Load balancing method of service cluster based on mean-variance |
| |
| Abstract: | When a large number of concurrent requests are allocated,the load scheduling mechanism is to achieve the load balancing of nodes in the network by minimizing the response time and maximizing the utilization ratio of nodes.In the load balancing algorithm based on genetic algorithm,the fitness function is designed to have an important influence on the load balancing efficiency.A service cluster load balancing method based on mean-variance was proposed to optimize the fitness function.The investment portfolio selection model mean-variance was used to minimize the response time,which was used to get the weight of each server's resource utilization,so as to obtain the optimal allocation combination.This method improves the accuracy and efficiency of the fitness function.Compared with other models in different service environment,the simulation results show that the load balancing algorithm makes the service cluster get a better balance performance in terms of node utilization and response time. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
|
