Abstract—Recently, cloud services and cloud computing have revolutionized both academic research and industrial practices. A corresponding focus on how to improve the performance of cloud computing is growing apace. It is a significant approach to allocate virtual machines (VMs) on a set of physical machines (PMs). Computing resources can be utilized effectively with the optimal distribution of the virtual machines among the physical machines. This study aims to establish the dynamic placement model of VMs by multi-objective programming (MOP) for minimizing energy consumption, maximizing effectiveness of physical machine, and minimizing the task waiting time. The genetic algorithm (GA) is used to solve the multi-objective programming models and compared with the greedy method (GM). Experiments are implemented to verify the effectiveness of the proposed methods.

Index Terms—Key words: Virtual machine placement, cloud computing, multi-objective programming, genetic algorithm, greedy method.

The authors are with Department of Computer Science and Information Engineering, National Dong Hwa University. No. 1, Sec. 2, Da-Hsueh Rd., Shoufeng, Hualien 97401, Taiwan.

Cite:Hanying Kao, Yumin Yang, "Dynamic Virtual Machines Placement in a Cloud Environment by Multi-Objective Programming," International Journal of Computer and Communication Engineering vol. 6, no. 3, pp. 193-200, 2017.