Scaling PostgreSQL to power 800 million ChatGPT users
OpenAI runs ChatGPT on a single-primary Postgres with nearly 50 global read replicas. Reads are offloaded, write-heavy shardable workloads move to Cosmos DB, and PgBouncer, cache leasing and multi-layer rate limits keep the primary safe.
- Published
- 2026-01-22
- Scale
- 800 million users, millions of QPS, ~50 read replicas
- Techniques
- read replicas connection pooling caching rate limiting workload isolation query optimization
- Stack
- PostgreSQL Azure PgBouncer Kubernetes Cosmos DB
Try this stack yourself
- Railway · Deploy something like this on Railway: spin up PostgreSQL plus your app in a few clicks. New accounts get $20 in credit (applied when you pick a paid plan).
Railway isn't part of OpenAI's setup in this story. It's just an easy place to try the same building blocks. - Better Stack · Monitor uptime and logs with Better Stack, and get alerted fast if a migration or traffic spike goes sideways on your own system.
Better Stack isn't part of OpenAI's setup in this story. It's a tool for watching your own systems while you try ideas like these.
Some links above are affiliate links: if you sign up, How They Scaled may earn a commission at no extra cost to you. We only list tools that fit the stack in the story.