Virtual machine consolidation for cloud data centers using parameter-based adaptive allocation

Abstract

Cloud computing enables cloud providers to offer computing infrastructure as a service (IaaS) in the form of virtual machines (VMs). Cloud management platforms automate the allocation of VMs to physical machines (PMs). An adaptive VM allocation policy is required to handle changes in the cloud environment and utilize the PMs efficiently In the literature, adaptive VM allocation is typically performed using either reservation-based or demand-based allocation. In this work, we have developed a parameter-based VM consolidation solution that aims to mitigate the issues with the reservation-based and demand-based solutions. This parameter-based VM consolidation exploits the range between demand-based and reservation-based finding VM to PM allocations that strike a delicate balance according to cloud providers' goals. Experiments conducted using CloudSim show how the proposed parameter-based solution gives a cloud provider the flexibility to manage the trade-off between utilization and other requirements.