摘要
随着云计算和虚拟化技术的发展,一种应用不再是只能部署在一台物理机上,而是可将物理机的计算资源虚拟化成一台虚拟机来部署该应用。传统的分配物理资源放置虚拟机,只考虑应用所需资源是否充足,而未考虑应用类型对物理资源需求的特殊性。本文提出计算资源适用性概念——计算资源当前状态适合部署的应用类型。当需要部署某一应用时,首先判别应用所属类型,同时获取应用对资源占用量;然后根据当前候选资源节点的监控数据,分别进行静态评估和动态评估,计算出综合评估结果,得出部署应用的虚拟机和物理机的映射关系。实验结果表明,与传统的虚拟机放置算法相比,提出的计算资源适用性评估算法,使得应用响应时间变短,提高了应用效率与资源利用率。
With development of cloud computing and virtualization, it is possible that one physical machine can be virtualized one virtual machine to support an application. In tradition, virtual machine placement just considers whether there is sufficient re- source, ignoring the relationship of application type and physical resource status. This paper gives the concept of compute re- source applicability, which means that the current status of computing resource is suitable for deploying which application type. When deploying an application, firstly we judge the application type and acquire the occupation of resource, then combine moni- tor data with the occupation to do static estimation and dynamic estimation. The optimal virtual machine placement policy can be achieved by compositing the results. The experimental results show that, the proposed solution could effectively improve the effi- ciency of application and the use ratio of resource, and shorten the response time of application.
出处
《计算机与现代化》
2014年第1期147-151,共5页
Computer and Modernization
关键词
云计算
虚拟化
虚拟机放置
应用类型
计算资源适用性
cloud computing
virtualization
virtual machine placement
application type
computing resource applicability
