Backup to several disks back to back

[mhdganji]

Hi,
I am using Version 9.5 and here is my question
I have a bunch of disks which are not included in a RAID array. Just a few disks attached to the system each of them let's say is 4TB
I want veeam job to backup my data and start writing it to first disk and if it's full move to the second disk and ...
This is what I think was possible in netbackup and it can help me a lot cause I have limitations in RAID hardware and software and RAID is not very important to me but in the same time I don't want to create several jobs and put each one's destination on one drive. As I said I like to say start backup here from this disk and if full go to next one .. all in one job
Any chance ?

Thanks in advance

[veremin]

Are you going to rotate those drives or something? If so, you can create "rotated drive" repository, insert the first disk, wait until it becomes full, then, insert the second one and so on.

If you're looking for a way to distribute load between drives equally, consider Scale-Out Backup Repository.

Thanks.

[mhdganji]

Thanks Eremin but I'm a little confused. Let me explain what I want in better way
Assume that I have four 2 TB disks and I have a job backing up ten VM's with a total backup size of 5 TB
I want to tell Veeam : Hey man, Store this backup in first Disk and when it is full go to next disk (2TB on first one, 2TB on second and 1 TB on the last one)
I donno and am not interested on how files are splitted (maybe 4GB parts, maybe 2tb parts or maybe I am able to define it) or this process is managed, I just like it to work and also be able to restore successfully
I cannot raid this disks by hardware (and I don't like to do that by windows because it's slow and error probable)

Now is this scenario possible,

Thanks a lot

[davide.depaoli]

Hello mhdganji,
If you don't care how files are splitted or written, I think the Scale-Out Repository option is what are you looking for. You can aggregate your 2TB 4 hdd into one biggest 8TB logical repository, but Veeam BR treats it as a single repository.

[mhdganji]

Thanks, I will test both of them. And BTW,
1- if one disk fails in Scaled-Out Repository, all the data will be lost ?
2- How VEEAM does that? writing files in all disks together or first on the first one, then on second and ...

[veremin]

What is the size of the biggest VM to backup? And what is the size of disk you're going to use?

[mhdganji]

I have about 10 VM's in my job. One of them (the biggest one) is 2.2 TB. Size of disks are 2TB (10 * 2 TB Disks)
As possible, I don't want to create different jobs and set each one to one disk because its management is difficult, disk space will not be utilized efficiently and ...

[veremin]

My suggestions would to:

1) Enable per-VM backup chain, so that each source VM goes into a separate backup file.
2) Create Scale-Out Backup and existing drives as its extents. Having Scale-Out Backup Repository should guarantee that available disk space is utilized efficiently.

Information regarding how backup server distributes load among extents can be found here.

Thanks.

[mhdganji]

Thanks I will check but any how is it possible for veeam to distribute a vm backup with size of 2.2 TB on these 2 TB disks (is it ok with per-vm backup?) it breaks the file into separate items ?

[veremin]

Nope, backup server does not split backup file into data chunks.

However, with per-VM chains the biggest single entity will be 2.2 TB VM. Typical data reduction ratio (compression and deduplication) is 50%. So, the size of backup file will be approximately 1 TB, which means that it can be perfectly accommodated by the said drives.

Thanks.

[mhdganji]

Oh no do not take me wrong.
The 2.2 TB is the size of backup file (done individually)
the machine itself is about 5.4 TB
Is there any chance yet ?

[veremin]

Ah, in this case I don't think you can avoid disk unification, using either hardware (RAID) or software means (Storage Spaces). Thanks.