Title: Deferred virtual machine migration

Authors: Xiaohong Zhang; Jianji Ren; Zhizhong Liu; Zongpu Jia

Addresses: College of Computer Science and Technology, Henan Polytechnic University, 2001 Century Avenue, Jiaozuo (454003), China ' College of Computer Science and Technology, Henan Polytechnic University, 2001 Century Avenue, Jiaozuo (454003), China ' College of Computer Science and Technology, Henan Polytechnic University, 2001 Century Avenue, Jiaozuo (454003), China ' College of Computer Science and Technology, Henan Polytechnic University, 2001 Century Avenue, Jiaozuo (454003), China

Abstract: The rapid growth in demand for cloud services has led to the establishment of huge virtual machines. Online virtual machine (VM) migration techniques offer cloud providers means to reduce power consumption while keeping quality of service. However, the efficiency of online migration is suboptimal since it is degraded by the execution of redundant migrations. To alleviate this problem, we introduce a load-aware VM migration technique. The key idea is to defer these migrations and perform a quick analysis of the loads on target servers before launching any migration. This consolidation is applied in two steps: first all migrations are tested and a set of candidates that are suspected to be redundant are formed and postponed for a short time. Then, the servers are analysed and only a subset of the migration candidates is activated. Our selection mechanism is conservative in the sense that it avoids selecting and activating VM migrations that are unlikely to cause a harmful overloading on the target servers. Taking this conservative migration policy leads to an overall effective execution of virtual machines. Our experiment results demonstrate the usefulness and effectiveness of our method.

Keywords: cloud computing; virtual machine migration; virtual machine consolidation; power consumption.

DOI: 10.1504/IJHPCN.2019.101260

International Journal of High Performance Computing and Networking, 2019 Vol.14 No.2, pp.229 - 236

Received: 24 Aug 2016
Accepted: 06 Jun 2017

Published online: 24 Jul 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article