Search-Based Performance Testing and Analysis for Microservice-Based Digital Power Applications

Modern digital power applications (e.g., power trading system) are moving toward microservice architecture, in which each service is developed and managed independently, and new features and updates are delivered continuously. It is essential to manifest the performance problem of the application early and often, before the code ships to the production. In this paper, we propose a novel performance testing and analysis approach for microservice based application, which first designs a multi-objective search technique to expose the performance problem, and then detects potential performance bottleneck by aggregating and comparing normal/abnormal traces. The evaluation results on an open source microservice benchmark and an industrial digital power application show that our approach is effective, and can expose performance anomaly and detect performance bottleneck effectively.