摘要
在分布式云计算系统中,数据密集型计算可导致数据中心间的数据调度,而合理的数据分布式存储可有效减少数据中心间的数据调度,提高用户的数据存储和计算效率。为此,建立数据中心间数据调度的数学模型,通过遗传算法的全局优化能力和迭代产生更好的近似解,得到数据存储的最佳逼近。试验结果表明,遗传算法可有效计算近似最优的数据分布式存储,并最大限度减少数据中心间的数据调度。
In distributed cloud computing systems,data-intensive computing can lead to data scheduling between data centers.Reasonable data distributed storage can effectively reduce data scheduling between data centers and improve user data storage and computing efficiency.A mathematical model for data scheduling between data centers was estabilised.Through the global optimization ability of the genetic algorithm,a better approximate solution was generated through iteration,and the best approximation of the data storage was finally obtained.The experimental results showed that the genetic algorithm could effectively calculate the approximate optimal data distributed storage,and minimized the data scheduling between data centers.
作者
党倩
李策
王刚
吴天宇
DANG Qian;LI Ce;WANG Gang;WU Tianyu(State Grid Gansu Provincial Electric Power Company Information and Communication Company,Lanzhou 730000,China;Gansu Tongxing Intelligent Technology Development Co.,Ltd.,Lanzhou 730000,China)
出处
《电工技术》
2018年第23期33-37,49,共6页
Electric Engineering
关键词
云计算
电力大数据
数据布局
遗传算法
数据调度
cloud computing
power big data
data layout
genetic algorithm
data scheduling
