Scale-Out Repository Design

Hi,

For the performance tier you can use either block or object storage. If using block you’ll want to use a Hardened Repository (running on Linux) to enable immutability. Typically for this you’ll just want a server with a lot of disks such as a Dell R740XD2 (if you need THAT many disks) to get maximum capacity on the performance tier. But depending on how much you want to keep on performance tier that might be overkill. Avoid NAS devices as they can’t be immutable when using block storage. If using object storage there are plenty of options on the market. Object First, Scality Artesca, VAST Data are all examples of good products in this space.

Capacity tier you say to “buy” but do you actually want to buy storage or use something cloud based? In all scenarios the storage will be object based so whether you want to use an object storage vendor like I mentioned in the performance tier or an S3 compatible service like Wasabi or even S3 or Azure Blob themselves.

Finally for archive, until recently this had to be Azure Archive or S3 Glacier/Deep Glacier, but with 12.1 some on-prem solutions that are s3 compatible with data tiering are now supported but there aren’t many in this space yet as qualifications efforts are continuing, reach out to your Veeam sales representative for up to date information on that one.

For Performance tier, I recommend Harden Linux repo based on SSD disks, for fast backup and recovery for short retention

Hardened Repository - User Guide for VMware vSphere (veeam.com)

For Capacity tier, I recommend Local object storage

Veeam Alliance Partner Integrations & Qualifications

For Archive tier, I recommend public cloud S3 archive object storage like: AWS S3 Glacier or Azure blob