Cloud computing has been focused by a wide range of researches and organizational considerations. One of the most important challenges in this area is to reduce the energy consumption of cloud data centers. Virtualization, which is one of the key features of the machine consolidation technique, is a major step to improve the energy consumption and resource management in this area. Selection of appropriate virtual machines for migration and consolidating them on fewer servers is one of the challenges of virtual machine consolidation process. In this paper, a new approach is introduced to distribute migrations among different categories of virtual machines based on their importance in order to improve the cloud services. By introducing a matrix of virtual machines and their migration rates, the proposed method of this paper has reduced the migration of virtual machines to reduce their service level agreement (SLA). The results of the research are evaluated using the CloudSim simulator and the validity of the assumptions is proved experimentally.
展开▼