一种电力云数据中心的任务调度策略

随着智能变电站一体化监控平台及通信信息平台的建设与发展,电力数据呈现爆炸式的增长,而现有电力数据中心在处理海量数据时效率不高,无法满足电力用户多服务质量(QoS)需求。通过对电力云数据中心设计方案的分析,研究了云数据中心的任务调度策略。结合电力数据中心自身的特点,定义了一种电力用户多QoS评价模型,给出一种基于负载均衡的贪心算法,并以QoS总体效应值为目标进行任务调度。最后,以变电设备状态评估实验中心为例,对其故障诊断、状态评估、检修维护等任务进行模拟调度,并使用CloudSim平台进行仿真测试。结果表明,该算法能最大限度地满足电力用户多QoS需求,有效提高电力数据中心运行的效率,达到更加优越的负载均衡效果。

A Task Scheduling Strategy of Cloud Data Center in Electric Power Corporation

With the construction and development of the integrated monitoring platform of intelligent substation and communication and information platform, the electric power data appears to show an explosive growth. But the existing data centers in electric power corporations suffer low efficiency and are unable to meet the requirements of multi-quality of service (QoS) of electric power users. A multi-QoS evaluation model for electric power users is defined by referring to the characteristics of the data center in an electric power corporation. And based on an analysis of the design scheme for the cloud data center in an electric power corporation and the research on task scheduling strategies of the cloud data center, a load balancing based greedy algorithm is put forward with the target of QoS utility value. Finally, with the substation device condition assessment experiment center as an example, simulation scheduling is made for its tasks of fault diagnosis, condition assessment and repair and maintenance prior to test on the CloudSim platform. The result shows that the algorithm proposed can best satisfy the multi-QoS requirements of electric power users, improve the operating efficiency of data center in electric power corporation, and achieve a better load balancing effect.