CAJ | 학술논문

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。
침대현유적운계산임무조도책략부고필단수거중심내부부재균형급평등간대각항임무적문제，연구료기우복무등급협의（ service level agreements ，SLA）적다운수거중심임무조도궤제，설계료상응적임무조도효익모형화임무준입공제책략，제출료기우SLA적최대화수익임무조도산법（ SLA-MPS산법），실현료재다개운수거중심간조도자원、우선처리긴급임무적동시보증운복무상이익최대화。재CloudSim상적실험증명，SLA-MPS산법능가쾌임무향응속도，강저운복무상위약솔병제고기수익。
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-Profit-by-Scheduling algorithm ( SLA-MPS) was proposed .Eventually the goal to schedule resources among multiple cloud data cen-ters 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 .