基于DEA的能耗感知虚拟机资源分配算法

发布时间：2018-05-04 08:15

本文选题：云计算 + 虚拟化　；参考：《小型微型计算机系统》2015年01期

【摘要】：针对云计算虚拟化中心高能耗问题,提出一种能耗感知的虚拟机资源分配算法.该算法建立能耗模型,利用DEA评估模型对虚拟机部署方案历史资源需求和能耗进行评估,将CPU,磁盘,内存和带宽等资源作为投入指标,CPU,内存和磁盘的利用率和能耗之比作为产出指标,在评估过程中获取虚拟机资源的最佳权重和同时达到最佳技术效率和最佳投入规模的虚拟机部署方案.资源分配算法按照DEA评估结果进行虚拟机部署,DEA有效的虚拟机根据评估前的方案进行配置,DEA弱有效的根据DEA前沿面阴影计算出最佳投入资源来进行配置.实验结果表明,该算法不仅降低虚拟机能耗,还提高了资源利用率,达到节能效果.
[Abstract]:In order to solve the problem of high energy consumption in cloud computing virtualization center, a resource allocation algorithm based on energy consumption awareness is proposed. This algorithm establishes the energy consumption model, uses the DEA evaluation model to evaluate the history resource requirement and energy consumption of the virtual machine deployment scheme. Resources such as memory and bandwidth are used as input indicators, and the ratio of memory to disk utilization and energy consumption is used as an output indicator. In the process of evaluation, the optimal weights of virtual machine resources and the best technical efficiency and optimal input scale are obtained. Resource allocation algorithm deploys virtual machines according to DEA evaluation results; the efficient virtual machines are configured according to the scheme before evaluation; DEA weakly efficient; the best input resources are calculated according to the shadow of DEA frontiers. The experimental results show that the algorithm not only reduces the energy consumption of virtual machine, but also improves the utilization of resources.
【作者单位】：上海理工大学光电信息与计算机工程学院;上海理工大学信息化办公室;
【基金】：国家自然科学基金项目(61170277)资助上海市教委科研创新重点项目(12zz137)资助上海市一流学科建设项目(S1201YLXK)资助上海市研究生创新基金项目(JWCXSL1202)资助
【分类号】：TP302

【参考文献】