Digital Transformation and the Role of Dynamic Tooling in Extracting Microservices from Existing Software Systems

The emergence of microservices architecture has promised to offer scalable, maintainable, evolvable, agile, and failure-resilient software products attracting software companies to migrate their legacy systems to a microservices architecture. However, the migration process imposes a technical challenge to effectively identify potential microservice candidates from medium and large monolithic codebases. In recent years, automatic and semi-automatic migration methods analysing the static and the dynamic behaviour of monoliths at runtime are getting a growing attention. Along with the methods, dynamic data collection tools are proposed. In this paper, we systematically review, compare, and contrast dynamic data collection tools using criteria that were developed from existing studies on monolithic to microservice migration. We identified eight dynamic data collection tools and analysed the features using four open source codebases from Github. The paper further presents an in-depth study of two fitting tools applying them over the selected codebases. Such analysis enables researchers and practitioners to make informed decisions in the selection of dynamic data collection tools to automatically/semi-automatically migrate legacy codebases.