An efficient request-based virtual machine placement algorithm for cloud computing

Abstract

The energy efficiency of cloud computing has drawn gigantic attention due to the explosive growth of cloud services. Moreover, this growth extends the capacity of various resources of the datacenters. As a circumstance, the amount of carbon footprints generated from the datacenters is sharply increased. Therefore, the objective is to use the datacenter’s resources proficiently without compromising the user requirements such that energy consumption is minimized. The recent studies have shown that the user requirements are provided in the form of virtual machines (VMs) which are deployed in the physical machines (PMs) of the datacenters based on the resource utilization or decreasing order of the VM capacity. However, these studies have not considered the capacity of the user requests. In this paper, we propose a request-based VM placement (RVMP) algorithm by considering the capacity of the requests. The proposed algorithm assigns the user requests to the VMs and further assigns the used VMs to the PMs based on the capacity of the requests and VMs respectively. Our simulation results on five different datasets, which are generated using Monte Carlo method, show that RVMP improves performance in terms of the number of used VMs and PMs, average PM utilization and energy consumption of PMs compared to state-of-the-art algorithms.