Litcius/Paper detail

Efficient KPI Anomaly Detection Through Transfer Learning for Large-Scale Web Services

Shenglin Zhang, Zhenyu Zhong, Dongwen Li, Qiliang Fan, Yongqian Sun, Man Zhu, Yuzhi Zhang, Dan Pei, Jiyan Sun, Yinlong Liu, Hui Yang, Yongqiang Zou

2022IEEE Journal on Selected Areas in Communications43 citationsDOI

Abstract

Timely anomaly detection of key performance indicators (KPIs), <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g.</i> , service response time, error rate, is of utmost importance to Web services. Over the years, many unsupervised deep learning-based anomaly detection approaches have been proposed. To achieve good performance, they require a long period of KPI data for model training, which is not easy to guarantee with frequent service changes. Additionally, the training overhead is too significant for the vast number of KPIs in large-scale Web services. To address the problems, we propose an unsupervised KPI anomaly detection approach, named <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AnoTransfer</i> , by combining a novel Variational Auto-Encoder (VAE)-based KPI clustering algorithm with an adaptive transfer learning strategy. Extensive evaluation experiments using real-world data collected from several large-scale Web service providers demonstrate that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AnoTransfer</i> reduces the average initialization time by 65.71% and improves the training efficiency by 50.62 times, without significantly degrading anomaly detection accuracy.

Topics & Concepts

Computer scienceAnomaly detectionCluster analysisData miningInitializationArtificial intelligenceMachine learningWeb serviceScale (ratio)Overhead (engineering)World Wide WebPhysicsProgramming languageOperating systemQuantum mechanicsNetwork Security and Intrusion DetectionSoftware System Performance and ReliabilityAnomaly Detection Techniques and Applications
Efficient KPI Anomaly Detection Through Transfer Learning for Large-Scale Web Services | Litcius