A Time Series Prediction Approach Based on Hybrid Tuning for Database Performance Indicator in AIOps

一种基于综合调优的数据库性能趋势预测方法

¹ School of Computer, Northwestern Polytechnical University, Xi'an 710072, China
² Key Laboratory of Big Data Storage and Management, Ministry of Industry and Information Technology, Xi'an 710072, China
³ Bank of Communication, Shanghai 310115, China

Received: 5 November 2019

Abstract

One of the most important applications of the intelligent operation and maintenance of a cloud database is its trend prediction of key performance indicators (KPI), such as disk use, memory use, etc. We propose a method named AutoPA4DB (Auto Prophet and ARIMA for Database) to predict the trend of the KPIs of the cloud database based on the Prophet model and the ARIMA model. Our AutoPA4DB method includes data preprocessing, model building, parameter tuning and optimization. We employ the weighted MAPE coverage to measure its accuracy and use 6 industrial datasets including 10 KPIs to compare the AutoPA4DB method with other three time-series trend prediction algorithms. The experimental results show that our AutoPA4DB method performs best in predicting monotonic variation data, e.g.disk use trend prediction. But it is unstable in predicting oscillatory variation data; for example, it is acceptable in memory use trend prediction but has poor accuracy in predicting the number of database connection trends.

摘要

云数据库智能运维中的重要应用场景之一是对监控采集的大量性能时序数据进行趋势预测。提出一种基于Prophet模型和ARIMA模型的综合调优智能趋势预测方法AutoPA4DB（auto prophet and ARIMA for database）。该方法根据数据库性能监控数据的特征，进行了原始监控数据的预处理、预测模型自动调参和模型优化。采用加权的时序预测准确性度量WMC（weighted MAPE coverage），基于多个企业级数据库实例（包含10种性能指标）进行了实验验证。实验对比了5种不同时序模型的预测效果，结果表明在单调变化模式（如磁盘使用量）的数据中，文中提出的AutoPA4DB方法时序预测准确性最高；然而在震荡模式的数据中，预测效果不太稳定，例如内存使用率趋势预测效果较好，但数据库连接数趋势预测效果不理想。

1. Dang Y, Lin Q, Huang P. AIOps: Real-World Challenges and Research Innovations[C]//Proceedings of the 41st International Conference on Software Engineering: Companion Proceedings, 2019 [Google Scholar]
2. Pei Dan, Zhang Shenglin, Pei Changhua, et al. Intelligent Operation and Maintenance Based on Machine Learning[J]. Communications of the CCF, 2017, 13 (12): 68– 72 [Article] (in Chinese) [Google Scholar]
3. Lin Ziyu, Lai Yongxuan, Lin Chen, et al. Research on Cloud Databases[J]. Journal of Software, 2012, 23 (5): 1148– 1166 [Article] (in Chinese) [CrossRef] [Google Scholar]
4. Yuan Jidong, Wang Zhihai. Review of Time Series Representation and Classification Techniques[J]. Computer Science, 2015, 42 (3): 1– 7 [Article] (in Chinese) [Google Scholar]
5. Jan G., De Gooijer, Hyndman ROB J.. 25 Years of Time Series Forecasting[J]. International Journal of Forecasting, 2006, 22 (3): 443– 473 [Article] [CrossRef] [Google Scholar]
6. Yang Haimin, Pan Zhisong, Bai Wei. Review of Time Series Prediction Methods[J]. Computer Science, 2019, 46 (1): 21– 28 [Article] (in Chinese) [Google Scholar]
7. Taylor S J, Letham B. Forecasting at Scale[J]. The American Statistician, 2018, 72 (1): 37– 45 [Article] [CrossRef] [Google Scholar]
8. Han Dongming, Guo Fangzhou, Pan Jiacheng, et al. Visual Analysis for Anomaly Detection in Time-Series:a Survey[J]. Computer Research and Development, 2018, 55 (9): 1843– 1852 [Article] (in Chinese) [Google Scholar]
9. Cui Jiaxu, Yang Bo. Survey on Bayesian Aptimization Methodology and Applications[J]. Journal of Software, 2018, 29 (10): 3068– 3090 [Article] (in Chinese) [Google Scholar]
10. Rob J H, Anne B K, Ralph D S, et al. A State Space Framework for Automatic Forecasting Using Exponential Smoothing Methods[J]. International Journal of Forecasting, 2002, 18 (3): 439– 454 [Article] [CrossRef] [Google Scholar]