基于SLA的云计算多数据中心任务调度算法  

SLA-based Task Scheduling Algorithm among Multiple Data Centers in Cloud Computing

作　　者：于珊珊[1] 陈冬林[1] 李伟 鄂雪妮 

机构地区：[1]武汉理工大学电子商务与智能服务研究中心,湖北武汉430070 [2]新奇特车业服务股份有限公司,上海201824

出　　处：《武汉理工大学学报（信息与管理工程版）》2014年第3期345-349,共5页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国家自然科学基金资助项目(71172043);国家科技支撑计划基金资助项目(2013BAH13F01);中央高校基本科研业务费专项资金资助项目(2012-IB-063&2012-IB-060);武汉理工大学自主创新优秀博士论文培育基金资助项目(2013-YB-017)

摘　　要：针对现有的云计算任务调度策略仅考虑单数据中心内部负载均衡及平等看待各项任务的问题,研究了基于服务等级协议(service level agreements,SLA)的多云数据中心任务调度机制,设计了相应的任务调度效益模型和任务准入控制策略,提出了基于SLA的最大化收益任务调度算法(SLA-MPS算法),实现了在多个云数据中心间调度资源、优先处理紧急任务的同时保证云服务商利益最大化。在CloudSim上的实验证明,SLA-MPS算法能加快任务响应速度,降低云服务商违约率并提高其收益。The existing task scheduling strategy of cloud computing not only merely consideres the internal load balancing in a single data center , but also views all the tasks equally .To solve the two problems mentioned above , the mechanism of task scheduling among multiple cloud data centers was studied on the basis of service level agreements （ SLA ） .The corresponding benefit model of task scheduling and the strategies for task access control were designed ;and meanwhile the SLA-max-Profitby-Scheduling algorithm （ SLA-MPS） was proposed .Eventually the goal to schedule resources among multiple cloud data centers was achieved;the urgent tasks were treated firstly , and at the same time it ensured cloud providers the maximized profit .The test on CloudSim verified that SLA -MPS is able to accelerate the response time of the task , lower the default rate and improve the profit of cloud computing providers .

关 键 词：云计算 服务等级协议 多数据中心 利益最大化 任务调度算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]