Authors: Dipak Dabhi; Devendra Thakor
Addresses: Department of Computer Engineering, Chhotubhai Gopalbhai Patel Institute of Technology, Uka Tarsadia University, Bardoli, Surat, Gujarat, India ' Department of Computer Engineering, Chhotubhai Gopalbhai Patel Institute of Technology, Uka Tarsadia University, Bardoli, Surat, Gujarat, India
Abstract: The demand for cloud services has increased drastically in recent times. To fulfil the high demand for cloud services requires a higher number of servers of datacentres, and the datacentres consume high energy. Virtual Machine Consolidation (VMC) is a technique for lowering energy usage in datacentres by enhancing the utilisation of servers and shutting down underutilised server with maintain Service Level Agreement (SLA). Further, the VMC process is divided into four sub-policies: host overload detection, host underload detection, VM selection and VM placement. This article proposes the Hybrid VM Allocation and Placement (HVMAP) strategy, which uses migration control to detect the overload host and effectively locate the destination host. We utilised the CloudSim simulator with PlanetLab and Bitbrains data sets, to evaluate the performance. The evaluation results show that our strategy decreases Energy consumption and SLA Violation (ESLAV) by 101%, 54%, and 34% for the Planetlab data set, and 170%, 166% and 31% for the Bitbrain data set for three existing policies, respectively.
Keywords: cloud computing; VM consolidation; quality of service; service level agreement; VM selection; overload host detection; VM placement; underload host detection.
International Journal of Grid and Utility Computing, 2022 Vol.13 No.5, pp.459 - 471
Received: 14 Dec 2021
Accepted: 10 Feb 2022
Published online: 14 Oct 2022 *