Migrating Monoliths to Microservices based on the Analysis of Database Access Requests
Shang‐Pin Ma, Tsung-Wen Lu, Chung-Chieh Li · 2022
Many organizations seek to improve the maintainability and scalability of their computing systems by migrating from a conventional monolithic architecture to a microservice architecture (MSA). Note that service identification is the most crucial step in architecture migration. In the current study, we developed a microservice identification scheme based on analyzing database access to account for the importance of the pattern "Database Per Service" in microservices design. The proposed MIADA (Microservices Identification using Analysis for Database Access) system facilitates the clustering of service endpoints used in the identification and creation of microservices. The feasibility and efficacy of the proposed MIADA approach were evaluated using two existing service-oriented software projects, PlanApproval and CoCoME. The evaluation results show that the MIADA approach can effectively recommend appropriate clusters of service endpoints to construct microservices.