MorphingDB: A Task-Centric AI-Native DBMS for Model Management and Inference

TL;DR

MorphingDB introduces a task-centric AI-native DBMS embedded in PostgreSQL that automates model storage, selection, and inference. It advances in-database tensor handling with Mvec, a two-phase transfer-learning model selection, and DAG-based batch inference with pre-embedding/vector sharing to boost throughput. The system achieves favorable accuracy-resource-time trade-offs across nine public datasets (series, NLP, image) and shows competitive performance against both AI-native DBMSs and AutoML frameworks. Its architecture supports flexible storage modalities, scalable inference, and device-aware execution, enabling practical deployment of AI-driven analytics inside the database engine.

Abstract

The increasing demand for deep neural inference within database environments has driven the emergence of AI-native DBMSs. However, existing solutions either rely on model-centric designs requiring developers to manually select, configure, and maintain models, resulting in high development overhead, or adopt task-centric AutoML approaches with high computational costs and poor DBMS integration. We present MorphingDB, a task-centric AI-native DBMS that automates model storage, selection, and inference within PostgreSQL. To enable flexible, I/O-efficient storage of deep learning models, we first introduce specialized schemas and multi-dimensional tensor data types to support BLOB-based all-in-one and decoupled model storage. Then we design a transfer learning framework for model selection in two phases, which builds a transferability subspace via offline embedding of historical tasks and employs online projection through feature-aware mapping for real-time tasks. To further optimize inference throughput, we propose pre-embedding with vectoring sharing to eliminate redundant computations and DAG-based batch pipelines with cost-aware scheduling to minimize the inference time. Implemented as a PostgreSQL extension with LibTorch, MorphingDB outperforms AI-native DBMSs (EvaDB, Madlib, GaussML) and AutoML platforms (AutoGluon, AutoKeras, AutoSklearn) across nine public datasets, encompassing series, NLP, and image tasks. Our evaluation demonstrates a robust balance among accuracy, resource consumption, and time cost in model selection and significant gains in throughput and resource efficiency.

Figures (15)

• Figure 1: Overview of MorphingDB
• Figure 2: Model storage and management in MorphingDB
• Figure 3: Model selection based on task embeddings
• Figure 4: In-Database Vectorization Sharing
• Figure 5: The pipeline processing of batch inference