Neon
Cost
Performance
Pinned min/max CU (no scale-down)
Part of the Idle compute cost check · fix arrives as a guide
What it is
The compute endpoint has its minimum and maximum compute units pinned to the same value, preventing autoscaling.
Why it matters
You pay the maximum continuously rather than scaling with load. Pinning is usually done to avoid scale-up latency and ends up costing peak rates during every quiet hour.
How to fix it
Set a minimum matching your baseline and a maximum matching your peak, and let autoscaling move between them. Raise the minimum only if cold scaling proves to be a measurable problem.
Run them all on your app
Connect your repo and your live services with read-only scopes. The first scan is free, and nothing changes without your approval.