Domain Aligned Microservices Decomposition
Jay Gandhi, Raveendra Kumar Medicherla, Ravindra Naik · 2025
Refactoring a legacy monolithic application into microservices requires discovery of the current application's architecture, refactoring the architecture into microservices architecture, and then refactoring the application based on the new architecture.Architecture refactoring is a time consuming, expert driven process.In this paper, we propose an approach that recommends candidate microservices from a given monolith, aligning them with business services for the architect.Our approach first recovers the architecture of a given monolithic application using static analysis, then identifies the clusters of cohesive application fragments through unsupervised learning.These clusters are aligned with business services specified by a domain model.We have implemented our approach as a tool and compared the results with SOTA approaches on a suite of 4 java applications.Our tool recommended clusters are highly cohesive, loosely coupled, and better aligned with domain services.