Does machine learning for database optimization work in the real world?
Presented by:
Bohan Zhang
I am currently an engineer at OpenAI infra team. Previously, I co-founded OtterTune, an AI-powered database optimization startup. Before that, I worked with Prof.Andy Pavlo at the Carnegie Mellon database group on the OtterTune research project. I have spent several years in optimizing Postgres for many customers with both domain knowledge and machine learning.
I spoke at PostgreSQL Conferences in the past including PGConf Asia 2020, PGConf Asia 2021, PGConf Silicon Valley 2023 and Postgres Conference 2024.
Database Management Systems (DBMSs) are complex software that require precise tuning to achieve optimal performance on specific hardware and workloads. However, manual tuning by experienced administrators becomes impractical for large-scale DBMS deployments. To address this challenge, there has been a growing trend in both academia and industry to employ machine learning (ML) for automatic database optimization. OtterTune is a notable example of such an approach.
However, how effective is machine learning for database tuning, and does it work in real-world scenarios? In this talk, I will explore the challenges and insights gained from the OtterTune journey, covering both technical and business aspects.
- Date:
- 2024 November 6 10:10 PST
- Duration:
- 20 min
- Room:
- Ops: 421
- Conference:
- Seattle 2024
- Language:
- Track:
- Ops
- Difficulty:
- Easy