Affiliations: School of Big Data, Chongqing Vocational College of Transportation, Jiangjin, Chongqing, China
Correspondence:
[*]
Corresponding author. Chunrong Zhou, School of Big Data, Chongqing Vocational College of Transportation, Jiangjin 402247, Chongqing, China. E-mail: [email protected].
Abstract: Load balancing in cloud computing refers to dividing computing characteristics and workloads. Distributing resources among servers, networks, or computers enables enterprises to manage workload demands. This paper proposes a novel load-balancing method based on the Two-Level Particle Swarm Optimization (TLPSO). The proposed TLPSO-based load-balancing method can effectively solve the problem of dynamic load-balancing in cloud computing, as it can quickly and accurately adjust the computing resource distribution in order to optimize the system performance. The upper level aims to improve the population’s diversity and escape from the local optimum. The lower level enhances the rate of population convergence to the global optimum while obtaining feasible solutions. Moreover, the lower level optimizes the solution search process by increasing the convergence speed and improving the quality of solutions. According to the simulation results, TLPSO beats other methods regarding resource utilization, makespan, and average waiting time.
