Abstract: For enhancement of energy conservation and resource utilization nature of the green cloud computing, an analytical hierarchy process based virtual machine migration evaluation model is proposed. Such model utilizes the attribute clustering based collaborative filtering to rate attribute space of virtual machines for the decision making of best strategy of green virtual?machine?migration. By constructing pair-wise comparison matrix and adopting five-point rating scale, the model is able to evaluate the liability of migration and resource usage of virtual machine. Experimental result shows that the proposed model and method not only can significantly improve the resource utilization rate but also save the energy?consumption of cloud server cluster. It has the vital significance to the research work of green cloud computing.

Key words: green cloud computing, virtual machine migration, analytical hierarchy process, collaborative filtering

摘要： 在以节约能源和提高能源利用率为目的的绿色云计算背景下，提出一种基于层次分析法的虚拟机迁移评估模型。该模型通过基于属性聚类的协同过滤方法，针对虚拟机的属性空间进行评分，给出了最佳绿色虚拟机迁移策略的决策方法。通过建立虚拟机任务资源使用率指标体系及两两比较矩阵，采纳五级评分制对虚拟机迁移及任务资源占用进行了评估。实验结果表明，提出的模型和方法能够显著提高云计算集群的能源利用率、节约能源的消耗。该方法对绿色云计算的研究具有重要意义。

关键词: 绿色云计算, 虚拟机迁移, 层次分析法, 协同过滤