CLASSIFICATION OF DATABASE DESIGN PROBLEMS IN MICROSERVICE ARCHITECTURE AND RECOMMENDATIONS FOR THEIR RESOLUTION

Abstract

The paper investigates the problems of designing data storage systems in modern microservice architectures. The relevance of the study is driven by the widespread adoption of microservices and the Polyglot Persistence approach, which improve scalability and flexibility but significantly complicate data organization, integration, and consistency management. These challenges are particularly evident in HoReCa information systems, where intensive transaction processing, continuous data updates, integration with external services, and real-time operations require reliable and efficient data storage solutions. The aim of the study is to perform comprehensive analysis, systematization, and multi-level classification of data storage system design problems in microservice architecture and to develop practical recommendations for addressing them using modern architectural approaches and design patterns. The research is based on system analysis, comparative analysis of contemporary architectural solutions, and generalization of scientific publications and industrial practices related to microservice-based information systems. A novel three-level classification of data storage design problems is proposed, covering conceptual, architectural, and operational levels. For each level, the main challenges, their causes, representative examples from HoReCa, e-commerce, and banking information systems, and corresponding practical recommendations are presented. The study demonstrates that the effectiveness of data storage organization depends not only on selecting appropriate database technologies but also on defining bounded contexts, establishing clear Data Ownership principles, choosing suitable consistency models, integrating heterogeneous storage technologies, and organizing operational support for distributed systems. Attention is paid to the application of architectural patterns such as Database per Service, Saga, CQRS, Event Sourcing, Transactional Outbox, Transactional Inbox, API Composition, and Domain Events as practical mechanisms for solving the identified problems. The scientific novelty of the study lies in the systematization of data management problems in microservice architecture by dividing them into conceptual, architectural, and operational levels, determining cause-and-effect relationships between levels, and establishing correspondence between the identified problems, architectural patterns, conditions for their application, and accompanying constraints. The proposed approach can be applied in the development of scalable microservice-based information systems requiring reliable, consistent, and efficient data management.