Affiliations: [a] Institute of Applied Mathematics, Qujing Normal University, Qujing, Yunnan, China | [b] Colleg of Computer Science and Engineering, Qujing Normal University, Qujing, Yunnan, China
Correspondence:
[*]
Corresponding author: Jun Liu, Institute of Applied Mathematics, Qujing Normal University, Qujing, Yunnan, China. E-mail: [email protected].
Abstract: Virtual machine (VM) deployment has been a hot and difficult research topic in the cloud computing field. We address the problem of VM deployment in cloud computing where the situation of physical machine is considered, Traditional VM deployment algorithm considered the resource usage of task without considering the resource utilization of physical machine. There is a significant gap between the solutions obtained by existing algorithms and the optimal solutions, leading to lower resource utilization and unfair resource allocation. We propose an online task scheduling algorithm to solve VM deployment problem. Our proposed algorithm is based on greedy strategy and uses the jump tables to decide VM scheduling. Furthermore, our proposed algorithm considers the characteristics of the task, which will be divided into four types of tasks. This can math task’s property and physical machine to improve deployment efficiency. Experimental results demonstrate that our proposed algorithm is very suitable for cloud computing, effectively utilize resources, and improve performance of the system.
