Monitoring Integration Traffic in Microservices Architecture

Abstract

Microservice architecture has become a widely adopted approach for building modern software systems, offering advantages such as improved scalability, maintainability, and flexibility over monolithic architectures. However, monitoring microservices remains a complex task due to their distributed nature and the intricate interactions between independent services. This thesis investigates how integration traffic in Java-based microservices can be monitored more effectively. It identifies core principles for monitoring such traffic, examines the limitations of current solutions, and defines the key requirements for tools that support observability in microservice environments. A mixed-methods approach was applied, combining literature review with semi-structured interviews of industry professionals. The study’s main contribution is a set of design criteria for integration traffic monitoring tools that address both technical and organizational needs. The results show that existing systems often lack the detail and context required to trace data across services. Effective monitoring requires human-readable logs, standardized formatting, alerting mechanisms, intuitive visualization, and message-level traceability