Why Postgres Is Better Than MySQL for Automated Knob Tuning
summary
OtterTune deploys an agent that collects runtime metrics and current settings, trains ML models, and iteratively applies new configs for AWS Postgres and MySQL instances, claiming up to 3x performance differences from tuning. Postgres 14 exposes rich metadata in pg_settings so pg_settings exposes units and ranges and supports session/user/database scope changes, while MySQL only adds partial info in Variables Info in v8. Extensions with low level hooks like extensions like Citus and TimescaleDB let OtterTune read extra knobs and stats (pg_stat_statements, HypoPG) and reduce integration work.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 1:11OtterTune & knob tuning
- 3:10OtterTune architecture
- 4:29System view exposes all knob info
- 8:17Knob changing flexibility
- 10:23Citus & TimescaleDB with OtterTune
related talks
20mDebate: Autotuning
48mDBtune: AI-driven performance tuning for all PostgreSQL flavors
48mChatGPT Ain’t Got $%@& On Me! Next Generation Automated Database Tuning
50mPostgreSQL Versus MySQL - Isn't a Database Just a Database
04mPostgres Tuning is saved by Sovenath Shaw
27mBuilding safety tooling for risk-free AI tuning of Postgres
33mTuning PostgreSQL to work even better
45mTuning Hardware and the Operating System: Benchmarking on PostgreSQL