摘要
利用三支决策的基本思想,提出了面向实时云任务的细粒度任务合并调度算法.其主要思想是利用客户提交的实时任务的截止时间来计算松弛时间,按照松弛时间将任务放入紧急调度队列、正常调度队列和松弛调度队列.结合虚拟机的实际负载情况,对这三个队列提交任务进行调度.与ETC、ESTC、MTC、ETC&MQS、ESTC&MQS算法进行对比,结果表明,所提出的算法在满足用户SLA的前提下,能够有效降低云能耗.
Based on the basic idea of three-way decision,a fine-grained task-merging scheduling algorithm for real-time cloud tasks was established. The main idea was to use the deadline of the real-time task submitted by the customer to calculate the slack time. Then,the task was divided into three queues by using slack time,namely emergency dispatch queue,normal dispatch queue,and relaxed dispatch queue. Finally,the three queues were submitted for scheduling according to the actual load of the virtual machine. Compared with ETC,ESTC,MTC,ETC&MQS,and ESTC&MQS,the experimental results showed that the proposed algorithm could effectively reduce the energy consumption of cloud on the premise of satisfying the user SLA.
作者
姜春茂
王凯旋
JIANG Chunmao;WANG Kaixuan(College of Computer Science and Information Engineering,Harbin Normal University,Harbin 150080,China)
出处
《郑州大学学报（理学版）》
CAS
北大核心
2019年第2期66-71,共6页
Journal of Zhengzhou University:Natural Science Edition
基金
国家自然科学基金项目(61402131)
哈尔滨科技人才项目(2014RFQXJ073
2016RAQXJ036)
关键词
三支决策
云计算
任务合并
松弛时间
三支队列
three-way decision
cloud computing
task consolidation
slack time
tri-queue
