Automatic and Generic Periodicity Adaptation for KPI Anomaly Detection

Key performance indicator (KPI) anomaly detection (AD) is critical to ensure service quality and reliability. Due to the effects of work days, off days, festivals, and business activities on user behavior, KPIs may exhibit different patterns within different days, which we call periodicity profiles of KPIs. However, existing KPI AD approaches have difficulties in adapting to diverse periodicity profiles due to the lack of generality. In this paper, we propose an automatic and generic framework called Period, which can accurately detect the periodicity profiles through daily subsequences clustering, and improve the performance of AD methods by robustly and automatically adapting to different periodicity profiles. In our evaluation using several real-world KPIs with different periodicity profiles from large Internet-based services, the clustering algorithm used to detect periodicity can achieve about 0.95 accuracy on average. More importantly, further evaluation on 56 KPIs shows that Period can significantly improve the best ${F}$ -score of several widely used AD approaches by up to 0.66.

[1]  Xin Huang,et al.  Robust and Rapid Adaption for Concept Drift in Software System Anomaly Detection , 2018, 2018 IEEE 29th International Symposium on Software Reliability Engineering (ISSRE).

[2]  Samuel Kounev,et al.  Telescope: A Hybrid Forecast Method for Univariate Time Series , 2017 .

[3]  Eamonn J. Keogh,et al.  Exact indexing of dynamic time warping , 2002, Knowledge and Information Systems.

[4]  T. Warren Liao,et al.  Clustering of time series data - a survey , 2005, Pattern Recognit..

[5]  Mohammed Al-Shalalfa,et al.  Efficient Periodicity Mining in Time Series Databases Using Suffix Trees , 2011, IEEE Transactions on Knowledge and Data Engineering.

[6]  Ying Wah Teh,et al.  Time-series clustering - A decade review , 2015, Inf. Syst..

[7]  Hujun Bao,et al.  Understanding the Power of Clause Learning , 2009, IJCAI.

[8]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[9]  Roman Kern,et al.  Robust Parameter-Free Season Length Detection in Time Series , 2019, ArXiv.

[10]  Daniel Massey,et al.  Argus: End-to-end service anomaly detection and localization from an ISP's point of view , 2012, 2012 Proceedings IEEE INFOCOM.

[11]  Ulrike von Luxburg,et al.  A tutorial on spectral clustering , 2007, Stat. Comput..

[12]  Christos Faloutsos,et al.  Fast Time Sequence Indexing for Arbitrary Lp Norms , 2000, VLDB.

[13]  Subutai Ahmad,et al.  Unsupervised real-time anomaly detection for streaming data , 2017, Neurocomputing.

[14]  P. Rousseeuw.  Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[15]  Walid G. Aref,et al.  Periodicity detection in time series databases , 2005, IEEE Transactions on Knowledge and Data Engineering.

[16]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[17]  Ratul Mahajan,et al.  A provider-side view of web search response time , 2013, SIGCOMM.

[18]  Saeed Amizadeh,et al.  Generic and Scalable Framework for Automated Time-series Anomaly Detection , 2015, KDD.

[19]  Yin Zhang,et al.  Detecting the performance impact of upgrades in large operational networks , 2010, SIGCOMM 2010.

[20]  Peter J. Rousseeuw,et al.  Finding Groups in Data: An Introduction to Cluster Analysis , 1990 .

[21]  Lovekesh Vig,et al.  LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection , 2016, ArXiv.

[22]  Ali S. Hadi,et al.  Finding Groups in Data: An Introduction to Chster Analysis , 1991 .

[23]  OwezarskiPhilippe,et al.  Online and Scalable Unsupervised Network Anomaly Detection Method , 2017 .

[24]  Philip S. Yu,et al.  On Periodicity Detection and Structural Periodic Similarity , 2005, SDM.

[25]  Valentino Constantinou,et al.  Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding , 2018, KDD.

[26]  David J. Ketchen,et al.  THE APPLICATION OF CLUSTER ANALYSIS IN STRATEGIC MANAGEMENT RESEARCH: AN ANALYSIS AND CRITIQUE , 1996 .

[27]  Wesam M. Ashour,et al.  A dynamic method for discovering density varied clusters , 2013 .

[28]  Zihui Ge,et al.  Crowdsourcing service-level network event monitoring , 2010, SIGCOMM '10.

[29]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[30]  Dan Pei,et al.  Opprentice: Towards Practical and Automatic Anomaly Detection Through Machine Learning , 2015, Internet Measurement Conference.

[31]  Hae-Sang Park,et al.  A simple and fast algorithm for K-medoids clustering , 2009, Expert Syst. Appl..

[32]  Yang Feng,et al.  Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications , 2018, WWW.

[33]  Gérard Govaert,et al.  Assessing a Mixture Model for Clustering with the Integrated Completed Likelihood , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[34]  Luis Gravano,et al.  k-Shape: Efficient and Accurate Clustering of Time Series , 2015, SIGMOD Conference.

[35]  Balachander Krishnamurthy,et al.  Sketch-based change detection: methods, evaluation, and applications , 2003, IMC '03.

[36]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[37]  Rob J. Hyndman,et al.  Forecasting time series with multiple seasonal patterns , 2008, Eur. J. Oper. Res..

[38]  Meir Toledano.  Fast automated detection of seasonal patterns in time series data without prior knowledge of seasonal periodicity , 2016 .

[39]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .