云计算中基于遗传算法的数据布局策略  

Data Layout Strategy Based on Genetic Algorithm in Cloud Computing

作　　者：覃伟荣 QIN Weirong(College of Resources and Environment,Qinzhou University,Qinzhou 535011)

机构地区：[1]钦州学院资源与环境学院,钦州535011

出　　处：《计算机与数字工程》2020年第3期534-539,共6页Computer & Digital Engineering

基　　金：广西重点科技研发项目(编号:2017AB43024)批文号桂科计字[2018]126号资助。

摘　　要：数据布局的合理性直接影响数据中心间的数据调度效率,进而提高对用户的数据采集效率。论文以数据中心之间数据调度为基础建立数学模型,利用分布式云计算技术处理用户的海量数据,并提供高性能计算资源和海量存储资源模式。在分布式云计算系统中,数据密集型计算可以有效处理数据中心之间的数据调度,通过遗传算法的全局优化能力产生最佳的近似解,并最终获得数据布局的最佳近似结果。实验结果表明,遗传算法可以有效地计算出最优数据布局的近似结果,并使数据中心之间的数据调度最小化。The rationality of data layout directly affects the efficiency of data scheduling between data centers,and then improves the efficiency of data collection for users. Based on the data scheduling between data centers,this paper establishes a mathematical model,uses distributed cloud computing technology to process users’ massive data,and provides high-performance computing resources and massive storage resources mode. In the distributed cloud computing system,data intensive computing can effectively deal with the data scheduling between data centers,generate the best approximate solution through the global optimization ability of genetic algorithm,and finally get the best approximate result of data layout. The experimental results show that the genetic algorithm can effectively calculate the approximate results of the optimal data layout and minimize the data scheduling between data centers.

关 键 词：云计算 数据布局 遗传算法 数据调度 

分 类 号：TN914.34[电子电信—通信与信息系统]