How should I host and scale FastAPI?

Hey there, I love the service and community. I’ve been reading the documentation and threads, but I’m struggling on how I should approach hosting my particular application on; the platform feels very much like the future and the right long-term choice, but I’m somewhat paralysed by choices and possibilities.

I am currently writing a mobile application that uses FastAPI as its source:
the app’s users simply GET data - I would like this to be as fast as possible, including if it scales big
other hardware PUT, POST, & DELETE every few minutes, to keep the data current

I would like to start small (and free) with a proof of concept, but with something that’s built with scalability and a setup that ensures the data remains consistent across instances, even if the app is rebuilt.

So my question is: How should I approach this?

I could create one Persistent Storage drive with SQLite, but would that scale? I can imagine reading could get slow with more users and instances.

I could build PostgreSQL into the application along with FastAPI. If it was rebuilt, I imagine the data would be reset?

Do I setup a PostgreSQL for each FastAPI instance and how would this be achieved in Python to keep all of the data consistent throughout regions?

Is there another approach I’ve missed?

If anyone has suggestions on best practice and methods for my use case, I would really appreciate your thoughts :slightly_smiling_face: