摘要
云计算是基于互联网发展起来的新型计算模式,对网际间各个节点的计算能力达到了充分利用;为了提高用户对云计算的使用效率,一般要为用户任务进行合理的全局资源调度和本地资源调度;对于全局资源调度,提出了一种改进的DAG图方法,充分考虑了任务撤销后对全局资源调度的影响;对于本地资源调度,提出了一种基于任务延迟的资源调度方法,充分考虑了立即调度不能执行时,如何合理地配置本地计算资源和等待时间的问题;实验结果表明,提出的方法能较好地提高全局资源调度和本地资源调度的效能。
Cloud computing is a new computing model based on internet development. In cloud, every node can be made full use with its calculation ability. In order to improve the efficiency of cloud computing, cloud resource should be scheduling globally and locally. For the global resource seheduling, an improved DAG diagram method was proposed in this paper. Task revoeation was fully considered in this meth- od. For local resource scheduling, a method based on task delay was proposed in this papei＇. When immediate operation can not be per- formed, local eomputing resources and waiting time will be considered. Experimental results show that these tow methods improves efficiency of global and local resource scheduling.
出处
《计算机测量与控制》
北大核心
2014年第2期499-502,共4页
Computer Measurement &Control
基金
河南省科技计划重点项目(102102210416)
关键词
云计算
DAG图
任务延迟
资源调度
cloud computing
DAG diagram
task delay
resource scheduling
作者简介
杨照峰（1978-），男，河南襄城人，讲师，硕士，主要从事计算机技术应用及网络安全方面的研究。
