VM Selection and Allocation Policy to Optimize VM Migration in Cloud Environment
Neha Garg1, Damanpreet Singh2, Major Singh Goraya3
1Neha Garg, Department of Computer Science and Engineering, Sant Longowal Institute of Engineering and Technology, Longowal, India.
2Damanpreet Singh, Department of Computer Science and Engineering, Sant Longowal Institute of Engineering and Technology, Longowal, India.
3Major Singh Goraya, Department of Computer Science and Engineering, Sant Longowal Institute of Engineering and Technology, Longowal, India.
Manuscript received on 14 March 2019 | Revised Manuscript received on 19 March 2019 | Manuscript published on 30 July 2019 | PP: 3444-3449 | Volume-8 Issue-2, July 2019 | Retrieval Number: B2700078219/19©BEIESP | DOI: 10.35940/ijrte.B2700.078219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Cloud computing, a metered based technology provides the services using virtualized technology over the internet. In the cloud environment, to improve the performance (such as utilization of the resources, energy minimization) extreme number of virtual machines (VMs) can be installed on the servers as per their resource capacity. In this way, servers can be overloaded. Overloaded servers consume more energythan normal status servers. VM migration (VMM) is an efficient technique to become a server in a normal state. VMM technique is used to consolidate the resources to increase resource utilization (RU) and reduceenergy usage. In the VMM technique, selection of VM such as which VM is migrated from one server to another server and allocation of VM on servers is an important aspect. Appropriate VM selection declines the numeral of VMMs and increasesenergy efficiency. Appropriate VM allocation declines the server to become overloaded. In this paper, the VM selection and allocation strategy is presented. CloudSim toolkit is used to verify the strength of proposed VM selection and allocation algorithm. Proposed VM Selection algorithm (MaMT) performs better than existing MiMT algorithm in terms of total energy consumption, number of hosts shut down, number of VMM, and average Service Level Agreement (SLA) violation rate. MaMT algorithm with resource aware provisioning (RAP) and MiMT+RAP algorithm combines both VM selection and allocation policies. RAP algorithm used both energy and RU parameters while allocating VM to the server.MaMTreduces the energy consumption up to 7.25% and reduces the SLA violation rate up-to 2.6% in comparison to MiMT algorithm. When VM selection and allocation policies combines together than more system performance is improved. MaMT+RAPreduces the energy consumption up to6.76% and reduces the SLA violation rate up-to 0.22% in comparison to MaMT algorithm.MiMT+RAPreduces the energy consumption up to15.23% and reduces the SLA violation rate up-to 0.95% in comparison to MiMT algorithm.
Index Terms: Cloud Computing, Energy Efficiency, Resource Provisioning, Resource Selection, Virtualization, VMM.
Scope of the Article: Cloud Computing