云环境下海量数据的分类优化调度方法研究  

A Classification Optimization Scheduling Method of Massive Data under Cloud Environment

作　　者：李玲[1] 邱芬[1] 

机构地区：[1]南昌大学科学技术学院,江西南昌330029

出　　处：《计算机仿真》2016年第5期315-317,430,共4页Computer Simulation

基　　金：江西省高校人文社会科学研究项目(TQ1210)

摘　　要：云计算环境下的海量数据具有不同的属性特征,使得调度时间与数据类型产生低关联性。传统的算法进行分类优化调度时,由于低关联性需要进行大量的迭代运算,使数据类型和任务调度时间满足调度要求,导致调度响应时间长,分类不准确的问题。提出基于预先分类算法的云计算环境下的海量数据的分类优化调度方法。通过分析云系统中海量数据周期任务可调度的必要条件和计算非周期任务的分布函数来组建云计算环境下的海量数据的任务调度模型,在促使任务处理器依据对任务的分类标准对任务的分派进行灵活的调整,同时计算各个调度任务对应资源等级与最小执行时间的乘积,利用乘积最小的值实现对云计算环境下的海量数据的优化调度。实验结果证明,基于预先分类算法的云计算环境下的海量数据的分类优化调度方法调度响应时间短,负载均衡。A classification optimization scheduling method of massive data under the cloud computing environment based on pre-classification algorithm is proposed.The schedulable necessary conditions of massive data periodic task in the cloud system are analyzed,and the distribution function of aperiodic task is calculated to form the task scheduling model of massive data under cloud computing environment.Then,the task processor is promoted,and based on the classification standard of tasks,the task assign is adjusted flexibly,meanwhile,the corresponding resource degree of each scheduling tasks is multiplied by the minimum execution time.Finally,the minimum value of product is used to realize the optimization scheduling of massive data under cloud computing environment.Experimental results prove that the proposed method has good performance with fast response and balanced load.

关 键 词：云计算环境 海量数据 调度模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]