[perrygeo]

> You can't scale an application that relies on a database heavily to this level because your fundamental constraint IS the database.

That's a big assumption. If your app is just an HTTP wrapper for SQL queries against a poorly-optimized database, sure.

But there are plenty of applications that spend time the majority of their time doing other stuff (http requests, FFI calls, g/cpu-intensive processing, etc.) where the database interaction is a small part of the overall performance profile - certainly not the bottleneck, even with hundreds of concurrent clients.

In those cases, rather than artificially throttling the concurrency to stay within your global max_connections, you would ideally run as many as you can (ie fully utilize your infrastructure) and grab database connections from an external pool when you need them.