Abstract:

ing at the problem of flat layout and complex structure in the virtual machine placement strategy, this paper adopted a virtual machine allocation framework based on package-cluster mapping to minimize the energy consumption of all physical machines. Based on this, an effective energy consumption model and an improved binary particle swarm optimization algorithm with adaptive weights were proposed to improve the speed and accuracy of package-cluster mapping framework. Experimental results showed that the improved binary particle swarm optimization algorithm was more superior to traditional binary particle swarm optimization in terms of convergence speed and optimization ability. Compared with the virtual machine allocation algorithm, the improved binary particle swarm allocation algorithm based on the clustering framework increased the CPU usage and effectively reduced energy consumption, which promoted the green energy saving.

Key words: Science and Technology, Shanghai 200093, China)