基于VOO方法的云计算平台多目标任务调度算法  

A Multi-objective Scheduling Algorithm of Many Tasks in Cloud Platforms Based on Method of VOO

作　　者：朱丽玲[1] 杨智应[1] 

机构地区：[1]上海海事大学信息工程学院,上海201306

出　　处：《计算机技术与发展》2017年第1期11-15,共5页Computer Technology and Development

基　　金：国家自然科学基金资助项目(61202021)

摘　　要：目前,云计算正不断兴起和发展,它作为一种新的技术和商业模式受到国内外学者的重视。而任务调度问题是云计算的核心问题之一,也是研究热点。针对云计算中的任务调度问题,以完成时间和货币成本作为任务调度的两个性能指标,基于POSH算法,主要采用序优化的方法对任务调度问题的解集进行优化,获得一个相对最优的解。实验中将多HEFT算法和POSH算法进行了对比,结果表明,提出的方法在完成时间波动相对小的情况下,能够降低货币成本。因此,所提出的算法是有效的。Nowadays, cloud computing, as a new technology and business mode,is constantly rising and developing. Scholars at home and abroad have paid great attention to it. However,the problem of task scheduling in cloud platforms is one of the core issues and it＇ s the hot spot in the area of research on cloud computing. For the scheduling of many tasks in cloud platforms that makespan and monetary costs as two performance metric, an improved algorithm is proposed based on POSH. The algorithm mainly uses method of VOO to optimize the set of task scheduling solution,then can get a semi-optimal good-enough solution from that. In the experiments, compared with HEFT and POSH,it is concluded Re proposed method can reduce the monetary cost of the scheduling under the fluctuation of makespan rela- tively small, which is effective.

关 键 词：云计算 向量序优化 任务调度 多目标 完成时间 货币成本 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]