Abstract: With the development of cloud computing,energy consumptionhas increased dramatically,which further limits the improvement of the overall performance of the cloud data center,and thus the energy consumption issue has attracted the attention of industry and academia.Meanwhile,traditional particle swarm optimization algorithm(PSO) is widely used to solve data center task scheduling problems,but it has the shortcomings of slow convergence and low accuracy,and it is easy to ignore the cluster energy consumption problem.A chaotic mapping adaptive particle swarm optimization algorithm based on opposition-based lear-ning(OAPSO) is proposed.Firstly,the initial population is generated by the method of opposition-based learning,which makes the particles more evenly distributed in the initial solution space and improves the quality of the initial population.Secondly,a nonlinear decreasing dynamic inertia weight stra-tegy is introduced into the particle updating mode to change the particle optimization ability,so as to balance the local search and global search and avoid the algorithm falling into the local optimal.Thirdly,the chaotic mapping strategy is introduced to generate new solutions by perturbation and mutation at the optimal location,which improves the ability of the algorithm to jump out of the local optimal.Finally,the proposed algorithm is verified by experiments on the Cloudsim platform,and the results show that,compared to PSO,OBL_ TP_PSO and SAPSO,OAPSO algorithm has higher resource utilization and better energy-saving effect.

Key words: Cloud data center, Task scheduling, Particle swarm optimization, Chaotic mapping, Energy consumption optimization