Towards Identifying Microservice Candidates from Business Rules Implemented in Stored Procedures

During the 1980 and 1990's decades, relational database management systems arose as an alternative to implement and store application business logic due to its robustness. Many of those legacy systems suffer from several problems such as low scalability, database vendor lock-in, and complex maintenance and evolution. With the success of lightweight virtualization techniques and new distributed architectures, mainly the microservices, companies are migrating legacy systems to this architectural style. Although several studies have proposed migration processes and reported migration experience to microservices, to the best of our knowledge, none of them has addressed systems whose business rules are implemented in database artifacts, particularly stored procedures. Therefore, this paper presents a process to identify microservice candidates from application business rules implemented in stored procedures. We applied the process to a real large scale system, for which 357 business rules were mapped and 13 microservices were identified. In addition, the process helped to find out many duplicated pieces of code, thus also improving the system maintainability.