基于多目标蚁群改进算法的虚拟机放置策略

Virtual machine(VM)placement is a key problem of cloud computing. Currently, many research focus on the management of data center. A lot of studies are targeted maintaining the performance properties ,as well as enhancing the utilization rate of system resource in early stage. In order to reduce power consumption, virtual machines can be moved to fewer physical nodes to achieve the purpose of saving power with the technology of migration. However, the optimization of these objectives are conflicting. Now there is little research take a comprehensive consideration about SLA , waste of resources and power consumption to optimize virtual machine placement problem， Most of the time only get a partial rather than a global optimal solution.In order to solve this problem ,this paper proposes a new virtual machine placement strategy based on Multi-Objective Ant Colony Improved(MACI) .Experimental results show that our algorithm can achieve the optimal balance in multiple conflict objectives,which effectively reduces the resource load and power consumption.