What is my certainty that this database of mine or other users will not crash one day, what is the guarantee that it will stay that way for good, the servers will not crash, what is it all about, what are these databases
Supabase gives you the tools, documentation, and community that makes managing databases, authentication, and backend infrastructure a lot less overwhelming.
Supabase by default uses AWS EC2 compute instances for the DB and a few other things. Then one of the AWS filesystem solutions and s3 for storage.
Other than that any solution you do should be backed up and you should have your own backups in addition to the providers backups if you are a serious business.
All databases and compute are subject to crashing. And backups or using read replicas if large enough to deal with failure are important.
Not sure if you mean Supabase by "your company". This is mainly a user helping user forum and not monitored regularly by Supabase.
Your instance is standalone in the AWS center as far as compute/disk/etc. Common things are the dashboard mainly and that is run on Vercel last I checked.
Not sure what you mean by Supabase on a server. Supabase is an enhanced Postgres Database, and other server components like PostgREST and Auth. That is all on an individual EC2 instance. Realtime/Storage API/Supavisor are on shared AWS servers around the world. The Dashboard is a react application running on Vercel.
If you depend on storage/realtime/supavisor(pooler) those are on shared AWS servers. Your biggest risk is software updates that you can't control that break your code. Auth and all the shared things I mentioned can be upgraded at anytime. You do not control more than your database version. This is similar to any SaaS provider that updates OS's etc like AWS.
Supabase is alot of parts. The website has nothing to do with the dashboard or database. The dashboard can crash and the database instances keeps running. You can see the history of issues here: https://status.supabase.com/
Yes. But now you throw in the words shared drives. Your EC2 CPU instance is a single point of failure. There is no fail over or redundancy unless you do read replicas. But that is an AWS thing and not dependent on Supabase engineers.