ASAD: Adaptive Seasonality Anomaly Detection Algorithm under Intricate KPI Profiles
Open Access
- 8 June 2022
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 12 (12) , 5855
Abstract
Anomaly detection is the foundation of intelligent operation and maintenance (O&M), and detection objects are evaluated by key performance indicators (KPIs). For almost all computer O&M systems, KPIs are usually the machine-level operating data. Moreover, these high-frequency KPIs show a non-Gaussian distribution and are hard to model, i.e., they are intricate KPI profiles. However, existing anomaly detection techniques are incapable of adapting to intricate KPI profiles. In order to enhance the performance under intricate KPI profiles, this study presents a seasonal adaptive KPI anomaly detection algorithm ASAD (Adaptive Seasonality Anomaly Detection). We also propose a new eBeats clustering algorithm and calendar-based correlation method to further reduce the detection time and error. Through experimental tests, our ASAD algorithm has the best overall performance compared to other KPI anomaly detection methods.Keywords
Funding Information
- National Natural Science Foundation of China (62072319, 2022YFG0041, 2021CDLZ-11, 6142103190415)
- Science and Technology Department of Sichuan Province (62072319, 2022YFG0041, 2021CDLZ-11, 6142103190415)
- Luzhou Science and Technology Innovation R&D Program (62072319, 2022YFG0041, 2021CDLZ-11, 6142103190415)
- Foundation of Science and Technology on Communication Security Laboratory (62072319, 2022YFG0041, 2021CDLZ-11, 6142103190415)
This publication has 26 references indexed in Scilit:
- BEATS: Blocks of Eigenvalues Algorithm for Time series SegmentationIEEE Transactions on Knowledge and Data Engineering, 2018
- Opprentice2015
- Generic and Scalable Framework for Automated Time-series Anomaly Detection2015
- k-Shape2015
- A provider-side view of web search response timeACM SIGCOMM Computer Communication Review, 2013
- The Move-Split-Merge Metric for Time SeriesIEEE Transactions on Knowledge and Data Engineering, 2012
- Experimental comparison of representation methods and distance measures for time series dataData Mining and Knowledge Discovery, 2012
- ARIMA Based Network Anomaly Detection2010
- SpADe: On Shape-based Pattern Detection in Streaming Time Series2007
- On Periodicity Detection and Structural Periodic Similarity2005