Am I correct in thinking that the cost of running an application
on fly.io is based on resources used and that the application
can have X number of instances running in parallell one
at least Edge location that is the closets to where the
request originates?
There is only one database running for all of the application instances?
That’s correct. You can run an instance in any location, and each one incurs a cost. Running a database in only one region isn’t recommended. If you decided to run something like our Multi-region Postgres, you’d want one replica per location. Each of those replicas would be billed as well. That said, most apps don’t need to run in more than a few regions!