Autotuning PostgreSQL: A deep dive into server parameter tuning with agentic AI

Manual PostgreSQL performance tuning is a high-risk, trial-and-error process that frequently leads to volatile performance, costly over-provisioning, and unnecessary downtime. For enterprises managing large, heterogeneous database fleets, these issues result in spiraling infrastructure costs and unpredictable behavior. This session introduces an agentic-AI approach to permanently solve this four-decade-old problem. We will detail the specific challenges and advantages of using adaptive AI for tuning complex workloads and share valuable lessons learned while developing a commercial autotuner, DBtune.

Learn how DBtune's intelligent agent analyzes unique workloads in real-time and dynamically adapts server parameters to maximize resource utilization and optimize query response time. We will provide a direct performance comparison between DBtune’s adaptive method and traditional, static tools like PGTune. Through a live demonstration, we’ll showcase DBtune operating across diverse environments—from self-hosted community PostgreSQL to managed cloud services like Amazon RDS, Aurora, Google Cloud SQL, and Azure Flexible Server.

The future of database tuning is AI-assisted.