A load balancing algorithm for virtual machines scheduling in cloud computing

Virtualization technology forms a variety of resources pools in cloud computing so that resources can be allocated on demand for users, but it makes resources management and resources allocation more complex and difficult. In view of the load balancing problem in virtual machine scheduling, an improved genetic algorithm is proposed in this paper. Through the improvement of selection and crossover operators in the genetic process, the optimization accuracy of the algorithm can be improved. We also propose a model of virtual machine load based on task execution time span. Compared with traditional genetic algorithm, FIFO and Round-robin scheduling algorithm, the simulation results show that our improved genetic algorithm performs better in the load balancing of the virtual machines.