An Efficient Green-Aware Architecture for Virtual Machine Migration in Sustainable Vehicular Clouds

Abstract

Virtual Machines (VMs) are used to increase flexibility and to reuse available resources in a vehicular cloud. These VMs allow the use of an environment capable of executing processes without the need for physical resources. Due to the mobility characteristics of vehicles in a vehicular cloud, these VMs can be migrated to different physical hosts. Furthermore, this migration of VMs can occur due to the need for load balancing, the scarcity of cloud resources, and the need to optimize the energy consumption of the vehicular cloud infrastructure. Establishing a mobile cloud migration policy becomes a major challenge due to the characteristics of the vehicular networks. In this paper, we propose an architecture for VMs migration in vehicular clouds based on the Energy-Aware paradigm. We also propose a new migration policy of VMs based on the energy expenditure of this migration, considering the energy expenditure of the VMs in the cloud and in the cloudlet, as well as the energy consumption of the communication between cloudlets, clouds, and vehicles. Simulation results show that the proposed policy achieved a reduction in the system's power consumption of about 10 percent. Furthermore, the proposed approach also reduced the number of drop migration of VMs by 5 percent.