PLRTune: Importance Pre-Sampling and LLM-Guided Reinforcement Learning for Automatic Database Tuning

2026-06-12

AI Summary

This work addresses key challenges in automated database tuning—namely, the high dimensionality of parameter spaces, substantial online evaluation costs, unreliable LLM or documentation-based hints, and limited optimization potential after initialization—by proposing a staged tuning framework. The approach first compresses the state space by identifying workload-relevant dominant parameters through importance-aware pre-sampling and re-ranking. It then refines documentation-derived hints using execution feedback to enhance initialization quality. Finally, it performs fine-grained tuning within the critical subspace via TD3 reinforcement learning. Integrating workload-aware parameter selection, execution-guided hint refinement, and reinforcement learning–based post-optimization, the method achieves an average performance improvement of 9.50% over the current state-of-the-art on MySQL and PostgreSQL, while reaching the best baseline’s peak performance 9.03× faster.

Abstract

Configuration tuning is critical to database performance, yet automatic database tuning remains challenging due to high-dimensional knob spaces, substantial online tuning cost, unreliable textual hints derived from Large Language Models (LLMs) or community documents, and the difficulty of exploiting the remaining optimization room after initialization. Hence, we propose PLRTune, a staged database tuning system that leverages workload-specific domain knowledge to identify a reduced search space and further optimize within this promising region. First, we develop an importance pre-sampling and reranking strategy to identify the dominant knob subset in a workload-specific manner and derive a compact state representation. Second, we design an execution-guided hint refinement technique to improve the initialization quality of documentation-guided tuning. Finally, we propose a post-tuning refinement stage that leverages Twin Delayed Deep Deterministic Policy Gradient (TD3) to explore the dominant knob subset and further exploit the remaining optimization room. We evaluate PLRTune on MySQL and PostgreSQL across diverse benchmark workloads. Compared with state-of-the-art approaches, PLRTune achieves the best final result on all tested workloads, improving over the corresponding best-performing alternative by 9.50% on average. Moreover, PLRTune reaches the strongest baseline's best performance level 9.03 times faster on average across workloads, demonstrating its practical runtime efficiency without sacrificing final tuning quality.

Problem

• automatic database tuning
• high-dimensional knob spaces
• online tuning cost
• unreliable textual hints
• remaining optimization room

Innovation

• importance pre-sampling
• LLM-guided tuning
• execution-guided hint refinement
• TD3 reinforcement learning
• database configuration tuning