GenAI-Augmented Polyglot Persistence for Microservices: Toward Self-Adaptive and Explainable Data Fabrics
Year: 2025
Type: preprint
Abstract: Microservices architectures have become the foundation for developing scalable and modern software systems, but they also bring significant challenges in managing heterogeneous and distributed data. The pragmatic solution is polyglot persistence, the deliberate use of several different database technologies adapted to a given microservice requirement - is one such strategy. This paper examines polyglot persistence in microservice based systems. This paper brings together theoretical concepts with evidence from practical implementations and comparative benchmarks of standard database platforms. A comparative framework is applied to relational, document, key-value, column-family and graph databases to assess scalability, consistency, query expressiveness, operational overhead and integration ease. Empirical data drawn from industry case studies such as Netflix, Uber, and Shopify, and survey data illustrate real-life adoption trends and challenges. These findings demonstrate that polyglot persistence increases adaptability , performance , domain alignment but also governance or operational complexity. To cope with such trade-offs, architectural patterns such as saga workflows, event sourcing, and outbox integration are discussed.
Source: arXiv (Cornell University)
Authors: Halili, Festim, Nuhiji, Anila, Veliu, Diellza Mustafai
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4417063423
Cites:
Cited by:
Related to: