Detection of Anomalies in Docker-Based Containers through Monitoring

Abstract

Microservices-based applications may be written and deployed in a cloud context considerably more quickly thanks to developing container technologies like Docker. For application providers, the reliability of these microservices becomes a top priority. Anomaly detection techniques can identify abnormal behavior that could result in unanticipated failures. In this study, a solution is created to monitor and analyze microservices’ real-time performance data to identify and treat anomalies. The component of the proposed solution is a container monitoring module that gathers container performance data, a data processing module using an anomaly detection algorithm, and an integrated fault injection module. The effectiveness of the proposed solution’s anomaly detection and diagnostics is also gauged through the utilisation of the fault injection module.