Asking because I'd like the make the jump myself and thought I'd first solicit ideas re: deployment/management from folks who have completed the journey.

Motivating factors are $, read replicas, and having a log drain to datadog.

My wish list for making the transition myself is:

- Hetzner or similarly cheap

- DB on volume

- Read replicas for batch jobs (don't need load balancing b/w them and master db)

- Backup

- Datadog agent

- Easy upgrade pathway -> some kind of IaC deploy of new Supabase version