@Michele Guion
Apologies for the delayed response. Please find the answers to your questions inline below:
What happens if the datacenter hosting the database has some kind of problems?
What happens will depend on the nature of the failure. If the datacenter is unreachable then your database will be offline. For these scenarios we recommend having a Disaster Recovery Plan in place to restore or failover your database.
Our backups are safe? Or are they stored in the same datacenter, so they could get corrupted?
By default, SQL Database and SQL Managed Instance store data in geo-redundant storage blobs that are replicated to a paired region. This helps to protect against outages impacting backup storage in the primary region and allow you to restore your server to a different region in the event of a disaster. More information on backups can be found here.
And what about point-in-restore (guaranteed for the last 35 days for Premium subscriptions)? Is this guaranteed also in case of issues or the data could be lost?
Same as above, by default backups are geo-redundant.
Are they safe in case of issues or should we configure a storage replica in another geographic location?
Locally redundant storage (LRS) replicates your data three times within a single data center in the primary region. LRS provides at least 99.999999999% (11 nines) durability of objects over a given year. When this durability is not enough you can consider implementing geo-redundancy. More information can be found here.
Hope this answers your questions. Let us know if you need further assistance.
-------------------------------
Please don’t forget to "Accept the answer" and “up-vote” wherever the information provided helps you, this can be beneficial to other community members.