Can SnapCenter with Oracle Plugin back up and restore DB to AWS S3?

[heightsnj]

either in the case of Oracle running on-prem, or running on AWS EC2?

Thanks for you advice in advance.

[Ontapforrum]

It's a great question, I have no idea how this works in the cloud but I am trying to start some conversation here. I am using generic term 'database' here instead of specific oracle plugin.  Hopefully, NetApp Cloud Gurus will chip in.

Considering the database Backups (Application consistent) : It has to be consistent volume block level. So, the primary backup has to be on either on-premise ONTAP system or CLOUD EC2/EBS/CVO (Primary Backup) and then perhaps you can vault the 'application consistent volume level' snapshot to Object-storage/S3 (Secondary Copy). I am guessing, this seems straightforward and should be doable with Snap Center. When it comes to restore, obviously depends upon the how far back you are restoring from  i.e either Primary EBS/CVO or secondary S3. I am guessing, if the files/data is needed from the S3 backup, then it must be restored to the Primary(Premise/CLOUD) and then bring the db up. Just as we use to do with normal vaulted backups, just my thoughts....

Something related I found: (Hope it makes sense)
https://cloud.netapp.com/blog/perform-database-backups-in-the-cloud-with-snapcenter
https://docs.netapp.com/us-en/occm/task_restore_backups.html#restoring-a-volume-from-a-backup-file

[heightsnj]

Are there documents that can fully explain how SnapCenter or it's plugins can back up databases to a object storage as a backup target, and restore them from the object storage?

[heightsnj]

@Ontapforrum 

I read two links you posted. What you basically suggested is to use SnapCenter with Oracle Plugin to back up Oracle DB, and then to use Cloud Backup to back up backups to S3.  Correct?

It would work, but it'd be better if SnapCenter can directly store backups into S3 without going through the other different tool. I guess, probably SnapCenter cannot do that now.

There is one the other thought, which is to use FabricPool, and auto policy, thus, I don't have to pay for Cloud Backup. However, it is unclear on the performance when restore comparing these two method: one is using Cloud Backup to restore, and the other is to restore upon on-read to pull data back from S3 due to FabricPool "auto" policy. Any thoughts?

[Ebin_SME_ORACLE · Accepted solution]

This feature is coming soon in Cloud Backup Service ( CBS). CBS can integrate with snapcenter seamlessly to copy oracle backups to s3 and perform restores . Please reach out to local NDA rep to sign you for the Beta program.