Volume 36, Number 2, April 2018
|339 - 344
|03 July 2018
Low-Energy-Orientated Resource Scheduling in Cloud Computing by Particle Swarm Optimization
School of Automation, Northwestern Polytechnical University, Xi'an 710072, China
In order to reduce the energy cost in cloud computing, this paper represents a novel energy-orientated resource scheduling method based on particle swarm optimization. The energy cost model in cloud computing environment is studied first. The optimization of energy cost is then considered as a multiobjective optimization problem, which generates the Pareto optimization set. To solve this multiobjective optimization problem, the particle swarm optimization is involved. The states of one particle consist of both the allocation plan for servers and the frequency plans on servers. Each particle in this algorithm obtains its Pareto local optimization. After the assembly of local optimizations, the algorithm generates the Pareto global optimization for one server plan. The final solution to our problem is the optimal one among all server plans. Experimental results show the good performance of the proposed method. Comparing with the widely-used Round robin scheduling method, the proposed method requires only 45.5% dynamic energy cost.
Key words: cloud computing / cost function / Pareto optimality / resource scheduling / particle swarm / scheduling alogorithms
关键字 : 云计算 / 代价函数 / 帕累托最优 / 资源调度 / 粒子群 / 调度算法
© 2018 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.