The current Veeam repository is
- Windows Server 2019
- DAS (2 x 36TB)
- NTFS format (large block sizes)
- Windows DeDuplication
Veeam is configured to store the data as "Dedupe-friendly".
The volume utilisations are
- Volume 1 : 164TB Veeam Backups using 14.5TB of disk space (2 per-vm backup job chains)
- Volume 2 : 152TB of Veeam Backups using 15.9TB of disk space (2 per-vm backup job chains)
This server is very old, and the choice of using Windows DeDuplication was due to their being no other option. ReFS was not a usable option at the time, and there hasn't been a way to move the vast amount of data off to a similarly sized server.
Until now and VeeaMover
So my question is: could VeeaMover migrate this data to a new server, changing from NTFS to ReFS and take advantage of block cloning between all the GFS recovery points? Without this, the disk space requirements would be ridiculously high.
Is there any tool to provide an estimate of the ReFS disk space required? A "WhatIf"? There are many factors going on: ReFS block cloning, Veeam compression changing from dedupe-friendly to something more aggressive, etc.
Even using VeeaMover to move to a new Windows Server with NTFS/DeDup isn't a viable option, due to the background nature of Windows DeDuplication: you would need to copy across the entire backup chain and that would need the full amount of disk space available until Windows DeDuplication runs.
The only other option is to start afresh on the new server and store the old server until the retention points have expired.
[Immutability is handled by Cloud storage SOBR, so not really looking at a hardened repository at this stage]
Thanks