[Feature Request] - Backup to Azure Blob via WAN Accelerators

[negrumanuel]

Hi all,

First post here - apologies if it's in the wrong location Or has been posted before - please move.

I would love to be able to perform backup copy jobs from on-prem to cloud, via the WAN acc svc, where the destination is Azure Blob.
The above setup works for example if the destination is an Azure file Share - this is our current setup. However that is more expensive than blob - so Ideally I would want to target blob.

I cannot afford to NOT go via the WAN acc svc because:
- The link between on-prem and cloud is low capacity
- As good a CBT driver might be, we still get fairly large incremental files. I cannot shift per night 700 GB of incrementals to cloud over my current pipe. WAN acc reduces that to 20-30 GB. Crazy.

Let me know your thoughts.
Thank you
Manuel.

[david.domask]

Hi Manuel,

I adjusted the topic title to include the proper name for WAN Accelerator.

Noted on the request, however this is by design as WAN Accelerators require heavy reads from the target repository in order to function properly, and this is regrettably not compatible with Hyperscalers like Azure. (The limitation does not apply to object storage added as S3 Compatible as those are typically on-premises and do not incur heavy fees / throttling like with Hyperscalers)

As a workaround, you might consider a staging repository in Azure instead of Azure File Share, and then do a Backup Copy from that Backup Copy; Backup Copy backup chains are independent of the source, so you could keep a short retention on the staging repository and then set desired retention on the 2nd Backup Copy; that would get you the benefits of the WAN Acceleration on 1st backup copy and the desired retention on the 2nd Backup Copy that resides in Azure Blob. It's still compute / storage usage, but I believe it ought be cheaper than Azure File Share, but I'm not immediately familiar with the costs.

[negrumanuel]

Thanks David. By repository in Azure - you mean a Disk attached to the VM? Because I did try to copy from the current Azure file share repo onwards to the blob but it's not finding anything at source. I did not try with a directly attached storage repo.

Thank you.

[david.domask]

Correct, if you place an actual VM with enough storage for a short-term retention and add that VM as a repository, you'll be able to utilize the Data Transfer settings. To reiterate, I understand this still means compute and storage costs (as well as networking), but you can keep the retention quite short as the backup copy chains will be independent of one another (i.e., the staging backup copy can have just a few restore points and the 2nd backup copy targeting Azure Blob will still copy most recent points while creating its own independent chain with whatever retention you desire)

[negrumanuel]

Thanks for the explanation. I'll give that a go. Happy to close this post.