NeurDB: An AI-powered Autonomous Data System
arXiv:2405.03924 · doi:10.1007/s11432-024-4125-9
Abstract
In the wake of rapid advancements in artificial intelligence (AI), we stand on the brink of a transformative leap in data systems. The imminent fusion of AI and DB (AIxDB) promises a new generation of data systems, which will relieve the burden on end-users across all industry sectors by featuring AI-enhanced functionalities, such as personalized and automated in-database AI-powered analytics, self-driving capabilities for improved system performance, etc. In this paper, we explore the evolution of data systems with a focus on deepening the fusion of AI and DB. We present NeurDB, an AI-powered autonomous data system designed to fully embrace AI design in each major system component and provide in-database AI-powered analytics. We outline the conceptual and architectural overview of NeurDB, discuss its design choices and key components, and report its current development and future plan.
References in corpus (19)
- Decision Transformer: Reinforcement Learning via Sequence Modeling
- Neo: A Learned Query Optimizer
- Hierarchical Representations for Efficient Architecture Search
- Bao: Learning to Steer Query Optimizers
- Privacy Preserving Vertical Federated Learning for Tree-based Models
- Database Meets Deep Learning: Challenges and Opportunities
- Balsa: Learning a Query Optimizer Without Expert Demonstrations
- Clipper: A Low-Latency Online Prediction Serving System
- Pruning neural networks without any data by iteratively conserving synaptic flow
- Towards Dynamic and Safe Configuration Tuning for Cloud Databases
- End-to-end Optimization of Machine Learning Prediction Queries
- Kepler: Robust Learning for Faster Parametric Query Optimization
- Model Slicing for Supporting Complex Analytics with Elastic Inference Cost and Resource Constraints
- Polyjuice: High-Performance Transactions via Learned Concurrency Control
- Unifying and Boosting Gradient-Based Training-Free Neural Architecture Search
- Hyper-Decision Transformer for Efficient Online Policy Adaptation
- MLCask: Efficient Management of Component Evolution in Collaborative Data Analytics Pipelines
- TRACER: A Framework for Facilitating Accurate and Interpretable Analytics for High Stakes Applications
- Powering In-Database Dynamic Model Slicing for Structured Data Analytics