A Business-Centric Approach to Automated Microservices Identification

Microservice-oriented architectures are becoming more and more popular in the academic and industrial domains as opposed to monolithic systems. With the help of microservices, huge, monolithic apps may be divided into more manageable, highly coherent, loosely linked pieces. Benefits like enhanced maintainability and scalability are provided by this. However, determining the optimal boundaries for these microservices continues to be a challenging task that might hinder the effective adoption of this design. This article presents an approach for automatically identifying microservices from Business Process (BP) using Natural Language Processing (NLP) methods and semantic analysis. The approach makes use of NLP to extract important data from business operations, providing insights into the connections between various tasks. As a result, discrete microservices are formed from activities that have comparable semantic properties, offering a data-driven method of breaking down the system. A case study centred on a bicycle rental system is carried out to exemplify and illustrate the approach.