一种基于优化QPSO的大数据云存储调度方法研究  

A BIG DATA CLOUD STORAGE SCHEDULING METHOD BASED ON OPTIMIZED QPSO

作　　者：于国龙 崔忠伟[1,2] 熊伟程 赵勇[1,3] 左羽 Yu Guolong;Cui Zhongwei;Xiong Weicheng;Zhao Yong;Zuo Yu(School of Mathematics and Computer Science,Guizhou Normal University,Guiyang 550018,Guizhou,China;Big Data Science and Intelligent Engineering Research Institute,Guizhou Normal University,Guiyang 550018,Guizhou,China;School of Information Engineering,Peking University Shenzhen Graduate School,Shenzhen 518000,Guangdong,China)

机构地区：[1]贵州师范学院数学与大数据学院,贵州贵阳550018 [2]贵州师范学院大数据科学与智能工程研究院,贵州贵阳550018 [3]北京大学深圳研究生院信息工程学院,广东深圳518000

出　　处：《计算机应用与软件》2021年第11期263-268,321,共7页Computer Applications and Software

基　　金：贵州省科学技术基金项目(黔科合基础[2016]1114号);国家科技部和国家自然科学基金奖励补助项目(黔科合平台人才[2017]5790-10);贵州省教育厅青年科技人才成长项目(黔教合KY字[2018]257);贵州省省级本科教学内容和课程体系改革项目(2017520065);贵州省省级重点学科“计算机科学与技术”(ZDXK[2018]007号);贵州省省级重点支持学科“计算机应用技术”(黔学位合字ZDXK[2016]20号)。

摘　　要：为了提升量子粒子群算法(QPSO)的全局收敛性及收敛效率,优化QPSO的全局最优位置计算方法。通过计算粒子当前适应度值与历史最高适应度值的差值占所有粒子适应度差值和的比重,来作为全局最优位置计算中各粒子局部最优位置的权重。粒子的寻优能力越强,其局部最优位置在全局最优位置计算中所占的权重就越高,使得粒子迅速向优秀粒子靠拢。将优化后的QPSO应用于大数据云存储平台的任务调度中。仿真实验表明,优化后的QPSO具有高效的全局搜索性能,能快速地为大数据云存储平台提供最佳任务调度策略。In order to improve the global convergence and convergence efficiency of the quantum particle swarm optimization(QPSO), the global optimal position calculation method of the QPSO is optimized. The difference between the current fitness value and the historical maximum fitness value of the particle was taken as the weight of the local optimal position of the particle in the global optimal position calculation. The higher the particle’s fitness is, the higher the weight of the local optimum position of particles is in the global optimal position calculation. The particle quickly converged to the best particle. The optimized QPSO was applied to the task scheduling of big data cloud storage platform. The simulation results show that the optimized QPSO has high efficiency in global search, and it can provide the optimal task scheduling strategy for big data cloud storage platform quickly.

关 键 词：QPSO 位置权重 云存储平台 任务调度 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]