GenAI-Augmented Polyglot Persistence for Microservices: Toward Self-Adaptive and Explainable Data Fabrics

Polyglot persistence lets microservices choose the best database for each workload, but today it is mostly static, manually configured, and hard to manage at scale. This paper presents GenAI-Augmented Polyglot Persistence (GAPP) — an architecture that uses generative AI and large language models (LLMs) inside the data orchestration layer.

GAPP automates workload-to-database selection, supports natural-language query federation, adapts orchestration based on runtime patterns, and provides explainable data lineage. Our prototype shows 94% query accuracy, complete lineage coverage, and an 85% reduction in development effort compared to traditional pipelines. The results highlight how AI can enable self-governing data fabrics that unify DevOps and DataOps. We conclude with key research challenges and a roadmap for building AI-native database management systems.