Anomaly detection and trend prediction are two fundamental tasks in automatic IT systems monitoring. In this paper, a joint model Anomaly Detector & Trend Predictor (ADTP) is proposed. In our design, the variational auto-encoder (VAE) and long short-term memory (LSTM) are joined together to address both anomaly detection and trend prediction. The prediction block (LSTM) takes clean input from the reconstructed time series by VAE, which makes it robust to anomalies and noise. In the mean time, VAE is able to fulfill the anomaly detection as only the normal statuses are encoded and decoded. In the whole processing pipeline, the spectral residual analysis is integrated with VAE and LSTM to boost the performance of both. The superior performance on two tasks is confirmed with the experiments on two challenging evaluation benchmarks.
View on arXiv