多维度资源环境下基于矢量代数模型的虚拟机部署  

VIRTUAL MACHINES PLACEMENT BASED ON VECTOR ALGEBRAIC MODEL IN MULTI-DIMENSIONAL RESOURCE ENVIRONMENT

作　　者：张扬 刘进[2] Zhang Yang;Liu Jin(Guangxi Vocational and Technical Institute of Industry,Nanning 530001,Guangxi,China;School of Computer and Electronics Information,Guangxi Univeristy,Nanning 530004,Guangxi,China)

机构地区：[1]广西工业职业技术学院,广西南宁530001 [2]广西大学计算机与电子信息学院,广西南宁530004

出　　处：《计算机应用与软件》2020年第5期9-14,29,共7页Computer Applications and Software

基　　金：国家自然科学基金项目(61661004);基于移动互联网的高等院校设备维护管理项目(KY2016LX574)。

摘　　要：多维度资源云数据中心环境下,资源利用率及系统能耗的同步优化是必须解决的问题。针对该问题,提出一种基于蚁群优化的虚拟机部署与合并算法。建立多维矢量装箱的虚拟机部署模型,定义模型优化目标;为了获取多维资源占用信息,设计基于矢量代数的多维资源利用模型;设计基于蚁群优化的虚拟机合并算法,通过信息素机制、启发信息、伪随机正比例规则及全局信息素更新机制,得到虚拟机部署的全局最优解。实验结果表明,比较同类型启发式算法,该算法在降低数据中心能耗与资源浪费方面均取得了更好的表现。In the multi-dimensional resource cloud data center environment,the synchronous optimization of resource utilization and system energy consumption is a problem that must be solved.We propose the virtual machine placement and consolidation algorithm based on ant colony optimization.We constructed the virtual machine placement model of multi-dimension vector binning,and defined the optimization objectives of the model.For obtaining the utilization information of multi-dimension resource,the multi-dimension resource utilization model was designed based on vector algebraic.We designed the virtual machine placement and consolidation algorithm based on ant colony optimization,and obtained the global optimal solution of virtual machined placement by the pheromone mechanism,heuristic information,pseudo-random proportional rule and global pheromone update mechanism.Experimental results show that compared with the same type of heuristics,our algorithm performs better in reducing the energy consumption and resource wastage of data centers.

关 键 词：虚拟机部署 虚拟机合并 蚁群算法 资源利用率 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]