부산대학교 도서관

Cloud datacenter consumes a lot of energy and there are significant opportunities to optimize the energy consumption. VM migration is one of the most renowned technique to migrate the VMs from the least used host and put it on sleep mode. However, this process also consumes a lot of energy and VM migration cost. Dynamic Voltage and Frequency Scaling (DVFS) can strength this process by giving only the required voltage to the running VMs on the host. In this paper, we proposed a hybrid approach for VM consolidation technique for energy optimization. The VM migration and DVFS decision applied for VM consolidation. A Workload-based Dynamic Threshold (WBDTH) approach design with a prediction technique to adjust the upper and lower threshold values according to the arrival rate. The proposed algorithm optimized energy consumption with a minimum Service Level Agreement (SLA) violations.