Optimization of Virtual Machine Layout in Cloud Data Center Based on Server Power Consumption Characteristics

Cloud data centers consume huge amounts of energy. More and more research focus on energy conservation and consumption reduction in data centers, and one of the important directions is the layout optimization of virtual machines running on servers. However, this paper finds that in the models and frameworks of existing research, almost no server power consumption characteristics are directly considered in the energy conservation problem. In fact, servers are the most important energy consumption units in the data center, of which the work characteristics and power consumption characteristics directly affect the energy consumption laws of other major energy consumption devices in the data center. Therefore, the real-time power consumption characteristics of the server should be emphatically considered in the energy conservation optimization problem. In this way, the best energy-saving effect can be achieved. Therefore, based on the characteristics of server power consumption, this paper establishes a cloud data center virtual machine layout optimization problem model, and takes the server cluster energy consumption directly as the goal of energy saving optimization. Then we use the Q-learning algorithm to solve the problem, and compares it with many popular optimization algorithms such as the best fitting descending method(BFD), particle swarm optimization(PSO), etc., in order to verify the effectiveness of the model. At the same time, it also proves that the model has good algorithm universality.