通信学报 ›› 2013, Vol. 34 ›› Issue (12): 33-41.doi: 10.3969/j.issn.1000-436X.2013.12.004

• 学术论文 • 上一篇    下一篇

基于CSP的能耗高效云计算资源调度模型与算法

林伟伟1,刘波2,朱良昌1,齐德昱1   

  1. 1 华南理工大学计算机科学与工程学院,广东 广州 510006
    2 华南师范大学计算机学院,广东 广州 510631
  • 出版日期:2013-12-25 发布日期:2017-07-05
  • 基金资助:
    国家自然科学基金资助项目;国家自然科学基金资助项目;广东省自然科学基金资助项目;广东省自然科学基金资助项目;广东省自然科学基金资助项目;中央高校基本科研业务费专项基金资助项目

CSP-based resource allocation model and algorithms for energy-efficient cloud computing

Wei-wei LIN1,Bo LIU2,Liang-chang ZHU1,De-yu QI1   

  1. 1 School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China
    2 College of Computer, South China Normal University, Guangzhou 510631, China
  • Online:2013-12-25 Published:2017-07-05
  • Supported by:
    The National Natural Science Foundation of China;The National Natural Science Foundation of China;The Natural Science Foundation of Guangdong Province;Guangdong Provincial Science and Technology Projects;Guangdong Provincial Science and Technology Projects;The Fundamental Research Funds for the Central Universities, SCUT

摘要:

利用约束满足问题对异构云数据中心的能耗优化资源调度问题建模,通过求解建立的约束模型可以获得能耗最优的资源分配方式,并在此基础上提出了能耗优化的资源分配算法 dynamicpower (DY)。与已有的算法MinPM、FFD、BFD相比,算法DY考虑了资源的异构性,能够降低云数据中心物理服务器的能耗。最后,利用Choco实现了提出的算法DY,并将DY与MinPM、FFD、BFD进行实验比较,实验结果表明,提出的算法在能耗上有明显优势。

关键词: 云计算, 资源分配, 能耗效率, 约束满足问题

Abstract:

The resource allocation for energy-efficient cloud computing in a heterogeneous environment was modeled as a constraint satisfaction problem (CSP).By solving the constraint satisfaction problem,the optimized allocation scheme minimizing energy consumption in virtualized cloud data centers was obtained.Based on the optimized allocation scheme,an energy-efficient resource allocation algorithm,dynamicpower (DY),which takes into account the heterogene-ity of resources,was proposed.The performance of algorithm was evaluated using Choco.Experimental results show that,compared with first-fit decreasing (FFD),best-fit decreasing (BFD) and minimizing the number of physical machines (MinPM),the proposed algorithm (DY) has less energy consumption.

Key words: cloud computing, resource allocation, energy efficiency, constraint satisfaction problem

No Suggested Reading articles found!