An Anomaly Detection Method for Stateful Stream Processing System
Guanghui Chang, Lu Zhao, Jun Liu, Peizhen Li · 2017
The SSPS (stateful stream processing system) is a distributed complex event processing system implemented by the framework of stream processing system.However, due to the nature of uncertainty, randomness and burstiness of the data stream and the complexity of complex event processing, the SSPS faces great challenges.In order to address the problem, this paper analyzes the problems of SSPS, on this basis, an anomaly detection method based on FKPCA-SVM is proposed.Experimental results show that the proposed method can efficiently and reliably detect SSPS, and ensure the system can operate normally.