云环境中基于弹性预留机制的虚拟资源供给策略  

Virtual resource provision policy based on elastic reservation mechanism in cloud environments

作　　者：屈喜龙[1,2] 肖鹏[1] 白泉[1] 

机构地区：[1]湖南工程学院计算机与通信系,湘潭411104 [2]华南理工大学计算机科学与工程学院,广州510641

出　　处：《系统工程理论与实践》2015年第6期1573-1581,共9页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(11172100);湖南省自然科学基金(13JJ9022);湖南省教育厅科学研究项目(13B015)

摘　　要：针对云计算环境中资源过度供给的问题,提出一种基于弹性预留机制的虚拟机供给策略.该策略在按需供给的基础上引入弹性预留机制,通过分析用户预留请求的统计特征,允许系统在一定条件下接纳用户的超额预留请求,从而在保证用户满意度的前提下提高云系统的资源有效利用率.理论分析给出了弹性预留机制的概率计算方法,实验分析则验证了所提资源供给策略的有效性以及关键策略参数对系统性能的影响.实验结果显示,与已有虚拟资源供给策略相比,本文所提策略能够有效提高云系统中的资源有效利用率,同时降低过度预留所导致的短期资源供给不足问题.To address the problem of resource overprovision in cloud environments,a virtual machine provision policy based on elastic reservation mechanism is proposed.It extends the conventional ＂ondemand provision＂ by introducing a novel elastic reservation mechanism,which allows cloud providers to accept user’s over-reservation requests under certain conditions so as to improve the resource effective utilization and user’s satisfaction.Theoretical analysis presents the approach to calculating the probability of elastic reservation,and experimental evaluation validates the effectiveness of the proposed policy and the effects of key parameter on system performance.The results show that comparing with existing techniques,elastic reservation based resource provision policy can significantly increase the resource effective utilization,and decrease the short-term resource under-provision caused by excessively reservations.

关 键 词：云计算 虚拟机 弹性预留 资源供给 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]