Best way to do archive in the cloud

[sakaufma1]

Hey all, hopefully easy question here. We're in the process of moving away from tape and to the cloud for offsite and archive. I just want to know the best way to do this. I'm assuming it will be a backup copy job with however many retention points we want and then check the box for "keep restore points as full backup for archival purposes". I'm thinking I'll keep a full per month. Then based on each job's retention, it will be kept mostly for a year, with some more important data (financial etc) being kept for up to 7.

I was planning on going through a Cloud Connect part for this but now that I'm seeing how many full monthly backup copies I'll have, I wonder if that's the best approach. Is it worth looking into AWS or Azure to dump the older backups to save money? How are most people handling this?

[Rick.Vanover]

Welcome to the Veeam community forums, Sakufma1.

The good news is you have a lot of options. The bad news is you have to choose which to use (and you can choose multiple!).

For the 7 years' worth of monthly full backups, just seeing that by itself; I would be inclined myself to have an arrangement like the following:

1) A Backup Job to a repository for short term (say 14 days?)
2) A Backup Copy job from that repository to a Scale-Out Backup Repository with the Capacity Tier configured to the object storage (AWS/Azure) you select with a month or so on-disk locally and backups older than 31 days kept in the capacity tier.

(Also consider some air-gap, such as tape backups or removable drives)

This will allow you to keep that long trail of backups in the cloud with some very intelligent metadata management. Meaning many objects that are the same as blocks on-premises will be kept in the cloud once.

Cloud Connect also is a good use case, especially if you consider the situation where you may need to restore to a new infrastructure, it is likely that the Cloud Connect service provider could provide you Infrastructure as a Service (IaaS) on-demand if needed.

[sakaufma1]

Are there limitations to what kind of AWS storage it can go to? If I'm putting just archive stuff out there, am I able to use the cheapest, "coldest" option, like AWS Glacier?

[Rick.Vanover]

For AWS, currently S3 is supported. There are gateway technologies that can go to Glacier, namely AWS VTL https://aws.amazon.com/storagegateway/vtl/

This may be an option as you can load VTL deep to Glacier this way, just note speed on restore the way Glacier works and set you expectations (SLA/RTO) accordingly.