[Objective] A container-based cloud computing center suffers from higher energy consumption, so an improved virtual migration strategy has been proposed to optimize energy consumption. [Methods] Firstly, we have studied the dependency among physical machines, virtual machines and containers of a cloud computing center in a container environment; secondly, analyzed the key factors affecting the energy consumption of the data center under such a dependency and established a mathematical model; finally, proposed an optimal utilization priority algorithm according to the non-linear relationship between utilization and energy consumption of different physical machines. [Result] The simulation experiments made on the ContainerCloudSim platform show that the method proposed in this paper can be used to effectively reduce the energy consumption of the data center in the container environment compared with traditional strategies such as random scheduling and maximum utilization. [Conclusion] In this paper, starting from the differences between different physical machines in the actual environment, the classical virtual migration strategy has been improved in order to reduce the energy consumption of the data center and moreover, the algorithm is simple, easy to implement and worthy of popularization.
展开▼