Improved sparrow algorithm based virtual machine placement

Abstract

Abstract To solve the problem of severe physical resource depletion in cloud data centers, a virtual machine placement method based on an improved sparrow search algorithm is proposed. Firstly, Chebyshev chaotic mapping and Levy flight disturbance strategy are integrated into the sparrow search algorithm to improve the allocation of population resources in the search space and increase the diversity of population search. In addition, a mapping encoding scheme is designed to map the virtual machine placement scheme into continuous position information, Finally, a cloud data center resource consumption model was established with the objective function of maximizing the utilization efficiency of physical machine resources. The simulation results show that this method has better performance in virtual machine migration and physical resource utilization, and the Service Level Agreement (SLA) violation rate is basically stable.