摘要
传统方法使用量子群遗传进化方法进行云存储系统任务调度的执行开销建模,在数据汇聚和协议传输中没有考虑量子态的相干性和感知节点的方向性,不能全局搜索最优量子位,执行开销不能实现最小化。提出一种基于量子群聚类的云存储调度最小执行开销建模算法,首先进行量子群聚类进化策略和云存储系统任务调度模型总体设计,设计基于量子群聚类的云存储系统任务调度分配协议,进行有效的资源调度设计,整合云计算中心资源,提高资源利用率,减少任务执行时间。仿真结果得出,该算法能使云存储系统任务调度执行开销与任务规模的匹配性能最佳,性能优于传统算法,在云存储信息管理系统等领域具有较好的应用价值。
Execution cost modeling using the traditional method of quantum group genetic evolutionary method for cloud storage system task scheduling, the coherence of quantum states and sensor nodes are not considered in the direction of data aggregation and transmission protocol, not global search optimal model, so it cannot realize the minimization of the execu- tion overhead. A minimum cost scheduling algorithm is proposed based on clustering of cloud storage quantum group, first quantum cluster evolution strategy and task scheduling model of cloud storage system overall design, scheduling tasks cloud storage system protocol of quantum groups is designed based on clustering, resource scheduling is designed effectively, the cloud computing center resources integrated, resource utilization rate is improved, the execution time of task is reduced. The simulation results show that, the algorithm can make the cloud storage system task scheduling performance overhead, and optimal matching task scale is obtained, performance is better than the traditional method, it has good application value in the field of cloud storage management information system.
出处
《科技通报》
北大核心
2015年第8期87-89,共3页
Bulletin of Science and Technology
关键词
云存储
构架
拓扑结果
任务调度
cloud storage
architecture
topology results
task scheduling