吸引力准则的仿生学在虚拟机迁移中的应用  

Application of bionics based on attraction criterion in virtual machine migration

作　　者：熊丽婷[1] 

机构地区：[1]南昌理工学院计算机信息工程学院,江西南昌330044

出　　处：《计算机工程与设计》2017年第12期3213-3217,3223,共6页Computer Engineering and Design

基　　金：江西省教育厅科学技术研究基金项目(151170)

摘　　要：为实现虚拟机迁移的能量效率最大化并提高资源利用水平,提出一种基于萤火虫优化(FFO)算法的快速虚拟机迁移方法,利用FFO收敛速度快和全局优化能力强的特点,通过萤火虫之间的吸引特性解决能量消耗问题。该方法由源节点的选择、虚拟机选择、目的节点的选择和距离更新值4个部分组成。将最大负载的虚拟机迁移到最小负载的活动节点,保持数据中心的性能和能量效率。仿真结果表明,与其它方法进行相比,该方法平均减少迁移达70%,能耗节约至少30%,提高了能量利用效率和资源利用水平。To maximize the energy efficiency and improve the level of resource utilization,a rapid virtual machine migration method based on fire fly optimization(FFO)was proposed,which made full use of the characteristics of fast convergence speed and global optimization ability of FFO,and the attraction characteristic between fire flies was adopted to solve the problem of energy consumption with energy efficiency optimization in the process of virtual machine migration.The proposed method consisted of four parts,the selection of source node,the selection of virtual machine,the selection of destination node and the distance update value.The virtual machine of maximum load was migrated to active nodes of minimum load,while maintaining the performance and energy efficiency of the data center.Results of simulation show that,the proposed method reduces the average migration by70%,and reduces the energy consumption by over 30%in comparison with other methods.The proposed method significantly improves the efficiency of energy utilization and resource utilization levels.

关 键 词：萤火虫算法 虚拟机迁移 能量消耗 云计算 负载 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]