计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (7): 68-73.DOI: 10.3778/j.issn.1002-8331.1509-0194

• 大数据与云计算 • 上一篇    下一篇

改进GWO优化SVM的云计算资源负载短期预测研究

徐达宇1,2,丁  帅2   

  1. 1.浙江农林大学 浙江省林业智能监测与信息技术研究重点实验室,杭州 311300
    2.合肥工业大学 计算机与网络系统研究所,合肥 230009
  • 出版日期:2017-04-01 发布日期:2017-04-01

Research on improved GWO-optimized SVM-based short-term load prediction for cloud computing

XU Dayu1, 2, DING Shuai2   

  1. 1.Zhejiang Provincial Key Laboratory of Forestry Intelligent Monitoring and Information Technology, Zhejiang A&F University, Hangzhou 311300, China
    2.Institute of Computer and Network Systems, Hefei University of Technology, Hefei 230009, China
  • Online:2017-04-01 Published:2017-04-01

摘要: 云计算资源负载短期预测是云计算平台实现资源高效管理和系统安全、稳定运行的重要前提和保障措施之一。为了其提高负载短期预测的预测精度,提出一种改进灰狼搜索算法优化支持向量机的短期云计算资源负载预测模型(EGWO-SVM)。首先介绍灰狼搜索算法(GWO)的基本原理;然后提出基于极值优化的改进GWO模型;最后根据最优参数建立短期资源负载预测模型,并通过仿真实验对EGWO-SVM的性能进行测试。实验结果表明,相对于参比模型,EGWO-SVM能更加准确地刻画云计算短期资源负载的复杂变化趋势,从而有效提升云计算资源负载短期预测的精度。

关键词: 云计算, 灰狼优化算法, 支持向量机, 极值优化, 预测

Abstract: Cloud computing resource load short-term forecasting plays an essential role in precondition and protection of highly effective resource management, secure and stable operation of system. Due to promoting the prediction accuracy of load short-term forecasting, EGWO-SVM is proposed. Firstly, the basic principles of Grey Wolf Optimizer(GWO) are presented. Then the improved extremal optimization based GWO model is proposed. Finally, the paper builds the prototype of resource load short-term prediction model according to the optimal parameters, and tests the performance of EGWO-SVM by simulation experiments. Experimental results show that compared with the reference models, EGWO-SVM can precisely characterize the complicated trends of cloud computing resource short-term load, efficiently promote the short-term resource load prediction accuracy in cloud computing.

Key words: cloud computing, Grey Wolf Optimizer(GWO), Support Vector Machine(SVM), extremal optimization, prediction