RDS Proxy: How Connection Pooling Actually Works
~10 min read
Why serverless/highly concurrent clients need pooling, and what RDS Proxy does differently from an application-level pool.
RDS Proxy: How Connection Pooling Actually Works is a Pro topic
Sign in, then upgrade to Pro or Power to unlock this topic and the full AWS curriculum.
Key points
- •Databases have a hard, instance-size-dependent cap on concurrent connections
- •Lambda's concurrency model can open many connections almost instantly, exhausting that cap
- •RDS Proxy pools and reuses a smaller set of actual database connections behind many client-facing logical ones
- •RDS Proxy also speeds up failover, since it manages reconnection to the new primary on the client's behalf