An Approach to Migrate a Monolith Database into Multi-Model Polyglot Persistence Based on Microservice Architecture: A Case Study for Mainframe Database

Abstract

Migration from a monolithic architecture to a microservice architecture is a complex challenge, which consists of issues such as microservices identification, code decomposition, commination between microservices, independent deployment, etc. One of the key issues is data storage adaptation to a microservice architecture. A monolithic architecture interacts with a single database, while in microservice architecture, data storage is decentralized, each microservice works independently and has its own private data storage. A viable option to fulfil different microservice persistence requirements is polyglot persistence, which is data storage technology selected according to the characteristics of each microservice need. This research aims to propose and evaluate the approach of monolith database migration into multi-model polyglot persistence based on microservice architecture. The novelty and relevance of the proposed approach are double, that is, it provides a general approach of how to conduct database migration from monolith architecture into a microservice architecture and allows the data model to be transformed into multi-model polyglot persistence. Migration from a mainframe monolith database to a multi-model polyglot persistence was performed as a proof-of-concept for the proposed migration approach. Quality attributes defined in the ISO/IEC 25012:2008 standard were used to evaluate and compare the data quality of the microservice with the multi-model polyglot persistence and the existing monolith mainframe database. Results of the research showed that the proposed approach can be used to conduct data storage migration from a monolith to microservice architecture and improve the quality of the consistency, understandability, availability, and portability attributes. Moreover, we expect that our results could inspire researchers and practitioners toward further work aimed to improve and automate the proposed approach.