云计算中基于多种群蚁群算法的虚拟机整合

VIRTUAL MACHINE INTEGRATION BASED ON MULTIPLE ANT COLONIES ALGORITHM FOR CLOUD COMPUTING

在云计算的发展研究中,数据中心的高能耗问题得到了广泛的关注,而虚拟机整合是解决数据中心高能耗问题的手段之一。其思想是通过将一些物理机上的虚拟机迁移到其他活跃的物理机上使得一些物理机切换到低能耗模式或睡眠模式,从而降低云数据中心的能耗。首次将多种群蚁群算法应用于虚拟机整合,提出基于多种群蚁群算法的虚拟机整合算法。该算法通过特定的目标函数寻找一个近似最优解。通过仿真实验验证了该算法在降低能量消耗和减少虚拟机迁移次数方面优于现存的两种较优的虚拟机整合算法。

In the development of cloud computing, high energy consumption of the data center has been widespread concerned, and the virtual machine integration is one of the means to solve the problem of high energy consumption. Virtual machine migration is a method of the number of virtual machines by the physical machine to migrate to make some of physical machines can sleep or enter low-power mode, which reduces power consumption of the cloud data center. In this paper, we present a multiple ant colonies algorithm to implement virtual machines consolidation. The proposed approach finds a near-optimal solution based on a specified objective function. It verifies through experiment that this algorithm outperforms existing virtual machines consolidation approaches in terms of energy consumption and the number of virtual machines migration.