To solve the resource scheduling problem in the large-scale cloud computing environment, this paper proposes an improved competitive particle swarm optimization algorithm (ICSO) to improve the efficiency of resource scheduling in cloud computing. Based on the multi-objective comprehensive evaluation model, firstly, the fitness function including task completion time, power consumption and load balance are established. Then, the more evenly distributed initialization particles are generated by the chaos optimization method, and the Gaussian mutation of adaptive probability is introduced to update the position of the victory particles, so as to improve the population diversity and enhance the global search ability. Simulation results show that under the same conditions, the algorithm can find the best scheduling scheme, which is suitable for large-scale resource scheduling, and the results are better than the comparison model.