Abstract
In order to support stable web-based applications and services, anomalies onthe IT performance status have to be detected timely. Moreover, the performancetrend across the time series should be predicted. In this paper, we proposeSeqVL (Sequential VAE-LSTM), a neural network model based on both VAE(Variational Auto-Encoder) and LSTM (Long Short-Term Memory). This work is thefirst attempt to integrate unsupervised anomaly detection and trend predictionunder one framework. Moreover, this model performs considerably better ondetection and prediction than VAE and LSTM work alone. On unsupervised anomalydetection, SeqVL achieves competitive experimental results compared with otherstate-of-the-art methods on public datasets. On trend prediction, SeqVLoutperforms several classic time series prediction models in the experiments ofthe public dataset.