Migration of Monolithic Systems to Microservices using AI: A Systematic Mapping Study

Abstract

The popularity of microservices architecture has been increasing considerably because of its capacity to alleviate monolithic architecture issues. Nonetheless, the migration of monolith systems to microservices is a complex task. This work aims to analyze and characterize the state of the art on migrating monolithic applications towards microservices (semi-) automatically using Artificial Intelligence (AI) techniques by applying a systematic mapping methodology. Results showed that clustering is the preferred IA technique to decompose a monolith, cited by 63% out of the 22 reviewed studies. Moreover, the most prevailing input type used in migration techniques was source code (36.4%).