Space Aware BGRU Microservice Fault Detection Algorithm

Microservice architecture is a new architecture pattern, which aims to provide users with more reliable, maintainable, and extensible software design services. However, with the continuous expansion of the scale of microservice application system, the proliferation of services and service interactions in the system make the system fault detection difficult. Detecting faults accurately and effectively is the key technology to ensure the system reliability and stability. From the perspective of microservice operation status and dependencies between services, this paper proposes a space-aware bidirectional gated recurrent unit (BGRU) microservice fault detection algorithm, which uses deep learning technology to mine hidden information that causes failures and combines space-aware attention to establish long-distance spatial dependency to improve the accuracy of model detection. The paper also conducts many experiments to demonstrate the effectiveness of the algorithm in microservice fault detection.