Neon Database Latency - Vercel Fluid Compute

Observe the latency querying Neon Serverless Postgres with different connection methods from Vercel Fluid Compute using the node runtime.

Data service

Waterfall

Executing complex API routes globally can be slow when the database is single-region, due to having multiple roundtrips to a single server that's distant from the user.

Samples

The number of samples to run for each location. A larger number of samples provides a clearer pattern of the average latency.