基于可用能力建模的云虚拟机动态调整策略  

A dynamic adjustment strategy for virtual machines in cloud based on availability capability modeling

作　　者：贾炅昊[1] 陈宁江[1] 李湘[1] 黄汝维[1] 

机构地区：[1]广西大学计算机与电子信息学院,广西南宁530004

出　　处：《广西大学学报（自然科学版）》2016年第3期796-803,共8页Journal of Guangxi University（Natural Science Edition）

基　　金：国家自然科学基金资助项目(61063012;61363003);国家科技支撑计划课题项目(2015BAH55F02)

摘　　要：云资源的可用性对资源管理带来一定的挑战,针对以往工作忽略云计算系统由于资源不足造成的不可用情景,提出一种可用性感知的虚拟资源动态调整策略(ADAR),利用马尔可夫模型对虚拟机可用能力进行建模,能够评估未来时刻资源的缺失情况对应用服务影响,根据评估结果对作为虚拟机资源进行调整指导。仿真实验验证了ADAR策略具有可行的效果,有效地降低资源浪费、降低云任务的完成时间和提高系统利用率。The availability of resources brings great challenge into resource management in cloud en- vironment. As for the ignorance of the impact of insufficient resourceson the availability of cloud systemin the precious research work, an availability-aware dynamic adjustment strategy for virtual machines, named ADAR, is proposed. It uses Markov model to characterize the availability capability of virtual machine to assess the impact of resource insufficiency on the application service in the future period, and makes corresponding adjustments according to the assessment. Simulation results provethe feasibility of ADAR strategy, which can effectively reduce the wasted resources and the makesoan of task. and improve the system utilization.

关 键 词：虚拟机 资源调整 可用性 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]