Best practice for large VM

[stevehathaway]

Q: My client has a new application server that is currently in the middle of a migration. The bulk of the data is files containing large images.
Currently the VM has a total of 40TB disk - one disk is 24TB.
This VM is expected to grow to over 100TB.
Most of this data is archival so will be largely static once the migration is complete.
Am I best to advise the application vendor to split the data across a number of Virtual disks; say around 25TB & then back these up using separate jobs rather than a single job?

[HannesK]

Hello,
one job is fine, but multiple disks are better for parallel processing. As 100 TB VMs require to span the server over multiple datastores, I would try to make it multiple VMs. But in general: try to make it as many disks "as possible".

Best regards,
Hannes

[Mildur]

Second to what Hannes has said, don't go over 64TB for a single Partition inside Windows.
Maximum allowed size for a partition is 64 TB for using VSS.

[stevehathaway]

Vielen Dank Hannes,

I'll work with the vendor to use the multiple drives - once the migration is done I'll add multiple jobs for this VM:
1 will only backup the O/S & live data (the archive disks will be excluded)
The others will backup just the archive drives probably on a weekly rotation

(The vendor supplied the server as an OVA & they set the initial drive with default allocation so it topped out at 16TB, so using 64K for the others )