Search-based many-criteria identification of microservices from legacy systems

The expensive maintenance of legacy systems lead companies to migrate such systems to a microservice architecture. This migration requires the identification of microservice candidates, which requires analysis of many criteria. Existing search-based approaches to solve this problem are only based on the coupling and cohesion criteria. To overcome these limitations, we propose a many-objective search-based approach for identifying microservice candidates. Its five objectives correspond to criteria pointed as useful by experienced developers. Our approach was evaluated in the context of a legacy system. The results show that our approach is very similar on optimizing the traditional criteria of coupling and cohesion, but much better when taking into account the additional criteria.