The Unreasonable Effectiveness of LLMs for Query Optimization

2024/11/05 by Peter Akioyamen, Akioyamen, Peter, Zixuan Yi +3 · 6 citations
Computer Science · #Advanced Database Systems and Queries #Cryptography and Data Security #Distributed and Parallel Computing Systems

paper · pdf · doi:10.48550/arxiv.2411.02862

Abstract

Recent work in database query optimization has used complex machine learning strategies, such as customized reinforcement learning schemes. Surprisingly, we show that LLM embeddings of query text contain useful semantic information for query optimization. Specifically, we show that a simple binary classifier deciding between alternative query plans, trained only on a small number of labeled embedded query vectors, can outperform existing heuristic systems. Although we only present some preliminary results, an LLM-powered query optimizer could provide significant benefits, both in terms of performance and simplicity.

Cited by

• Evaluating the Practical Effectiveness of LLM-Driven Index Tuning on Microsoft SQL Server
• This is Going to Sound Crazy, But What If We Used Large Language Models to Boost Automatic Database Tuning Algorithms By Leveraging Prior History? We Will Find Better Configurations More Quickly Than Retraining From Scratch!
• NeuSO: Neural Optimizer for Subgraph Queries
• SEFRQO: A Self-Evolving Fine-Tuned RAG-Based Query Optimizer
• MCTuner: Spatial Decomposition-Enhanced Database Tuning via LLM-Guided Exploration
• Bootstrapping Learned Cost Models with Synthetic SQL Queries

Related

• Training-Free Query Optimization via LLM-Based Plan Similarity
• LLM4Hint: Leveraging Large Language Models for Hint Recommendation in Offline Query Optimization
• A Query Optimization Method Utilizing Large Language Models
• LaPuda: LLM-Enabled Policy-Based Query Optimizer for Multi-modal Data
• Can Large Language Models Be Query Optimizer for Relational Databases?