Energy Efficient Virtual Machine Migration Algorithm

Journal of Engineering Research

Sa’ed Abed¹  Mohammad H AlShayeji¹  M.D. Samrajesh² 

[1] Computer Engineering Department, College of Computing Sciences and Engineering, Kuwait University,;Different Media

关键词: Cloud Computing;    Efficient Energy Conservation;    Load balancing;    Virtualization;    Virtual Machine;    Migration.;   

DOI  :  

学科分类：社会科学、人文和艺术（综合）

来源: Kuwait University * Academic Publication Council

【 摘 要 】

Today, a large volume of computing and data storage is done using cloud computing. Therefore, efficient datacenter resource utilization and energy consumption are considered important issues. Virtualization makes effective use of the datacenter’s hardware resources by using Virtual Machines (VM). VM(s) can function completely as a discrete unit and share the underlying hardware resources. The ability to migrate VMs within datacenter servers can considerably enhance the datacenter’s performance and resource utilization. However, most of the existing methods do not consider reducing energy consumption while migrating VM(s).In this paper, we present a novel Energy Efficient Virtual Machine Migration (EVM) technique that considers various vital factors of the datacenter servers while migrating VMs. Our proposed EVM technique is based on Energy based Server Selection (ESS) approach, which uses Highest Energy First (HEF) strategy for choosing the victim server to be switched off. In addition, ESS uses Lowest Energy First (LEF) strategy for target server selection to host the migrated VMs. The algorithm attempts to achieve efficient energy consumption at the datacenter by switching off underutilized servers. Our comparative evaluation results show that the proposed algorithm has lower overhead in terms of lower number of server state changes, VM migrations and oscillations, and yet outperforms existing methods. At 30% server load, the energy savings achieved using EVM is 31% more than Arbitrary Server Selection (ASS) and 15% more than First Fit Strategy (FFS). Moreover, EVM offers significant reduction in carbon footprint of 20% higher than ASS and 10% higher than FFS.

【 授权许可】

Unknown