通信学报 ›› 2021, Vol. 42 ›› Issue (1): 163-171.doi: 10.11959/j.issn.1000-436x.2021015

• 学术通信 • 上一篇    下一篇

基于ARIMA-RNN组合模型的云服务器老化预测方法

孟海宁1,2, 童新宇1, 石月开1, 朱磊1, 冯锴1, 黑新宏1   

  1. 1 西安理工大学计算机科学与工程学院,陕西 西安 710048
    2 陕西省网络计算与安全技术重点实验室,陕西 西安 710048
  • 修回日期:2020-11-22 出版日期:2021-01-25 发布日期:2021-01-01
  • 作者简介:孟海宁(1979- ),女,内蒙古乌海人,博士,西安理工大学副教授、硕士生导师,主要研究方向为云计算系统可靠性评估。
    童新宇(1996- ),男,陕西西安人,西安理工大学硕士生,主要研究方向为云计算系统性能预测。
    石月开(1995- ),男,陕西榆林人,西安理工大学硕士生,主要研究方向为云系统性能监控与故障诊断。
    朱磊(1983- ),男,陕西咸阳人,博士,西安理工大学讲师,主要研究方向为数据挖掘。
    冯锴(1997- ),男,内蒙古锡林浩特人,西安理工大学硕士生,主要研究方向为数据挖掘。
    黑新宏(1976- ),男,陕西延安人,博士,西安理工大学教授、博士生导师,主要研究方向为系统可靠性和安全性评估。
  • 基金资助:
    国家自然科学基金资助项目(61602375);国家自然科学基金资助项目(61773313);陕西省自然科学基础研究计划基金资助项目(2019JQ-749)

Cloud server aging prediction method based on hybrid model of auto-regressive integrated moving average and recurrent neural network

Haining MENG1,2, Xinyu TONG1, Yuekai SHI1, Lei ZHU1, Kai FENG1, Xinhong HEI1   

  1. 1 School of Computer Science and Engineering, Xi’an University of Technology, Xi’an 710048,China
    2 Shaanxi Key Lab Network Computer and Security Technology, Xi’an 710048, China
  • Revised:2020-11-22 Online:2021-01-25 Published:2021-01-01
  • Supported by:
    The National Natural Science Foundation of China(61602375);The National Natural Science Foundation of China(61773313)

摘要:

针对云服务器系统运行环境具有非线性、随机性和突发性的特点,提出了基于整合移动平均自回归和循环神经网络组合模型(ARIMA-RNN)的软件老化预测方法。首先,采用 ARIMA 模型对云服务器时间序列数据进行老化预测;然后,利用灰色关联度分析法计算时间序列数据的相关性,确定 RNN 模型的输入维度;最后,将ARIMA模型预测值和历史数据作为RNN模型的输入进行二次老化预测,从而克服了ARIMA模型对波动较大的时间序列数据预测精度较低的局限性。实验结果表明,ARIMA-RNN组合模型比ARIMA模型及RNN模型的预测精度高,且比RNN模型预测收敛速度快。

关键词: 软件老化, 云服务器, 预测方法, ARIMA模型, RNN模型

Abstract:

In view of the nonlinear, stochastic and sudden characteristics of operating environment of cloud server system, a software aging prediction method based on hybrid auto-regressive integrated moving average and recurrent neural network model (ARIMA-RNN) was proposed.Firstly, the ARIMA model performs software aging prediction of time series data in cloud server.Then the grey relation analysis method was used to calculate the correlation of the time series data to determine the input dimension of RNN model.Finally, the predicted value of ARIMA model and historical data were used as the input of RNN model for secondary aging prediction, which overcomes the limitation that ARIMA model has low prediction accuracy for time series data with large fluctuation.The experimental results show that the proposed ARIMA-RNN model has higher prediction accuracy than ARIMA model and RNN model, and has faster prediction convergence speed than RNN model.

Key words: software aging, cloud server, prediction method, auto-regressive integrated moving average model, recurrent neural network model

中图分类号: 

No Suggested Reading articles found!