Impact of Factors Affecting Pre-copy Virtual Machine Migration Technique for Cloud Computing

Abstract

Abstract Cloud managers trigger Virtual Machine (VM) migration to achieve the challenges of server consolidation, load balancing, resource management, and server fault situation. However, the performance of VM migration is affected by VM size, running application’s dirty rate and the number of migration iterations. Therefore, to measure VM migration performance by considering all these parameters becomes a challenging issue. The main contribution of this study is to evaluate the performance impact of these migration control parameters. Simulation results demonstrate that: (i) Migration time and downtime increases of two orders of magnitude while increasing the VM size, (ii) As the application dirty rate increases the migration time and downtime also increases, (iii) Half iterations of pre-copy migration are enough to reduce the migration time and downtime. In particular, the results obtained can help cloud service providers to find the performance impact of these factors on VM migration technique and meet the required goals of efficient deployment of cloud services.