Host-based backup of VMware vSphere VMs.
Post Reply
adam900331
Veteran
Posts: 312
Liked: 22 times
Joined: Dec 01, 2019 7:27 pm
Contact:

VM backup with large VMDK files

Post by adam900331 »

Hy!

Our customer has a VM which size is about 11Tb. There are several virtual disk with 1Tb size, but there is a disk with 6Tb size. I would like to backup with normal backup (snapshot). Do you have experience about the large size VM snapshot creation time? Are there any impact on VM side during the snapshot?

Thanks.
PetrM
Veeam Software
Posts: 3625
Liked: 608 times
Joined: Aug 28, 2013 8:23 am
Full Name: Petr Makarov
Location: Prague, Czech Republic
Contact:

Re: VM backup with large VMDK files

Post by PetrM »

Hello,

Basically, if the manual snapshot creation in vSphere client does not produce any issues, I wouldn't expect any problems with Veeam because snapshot creation/removing operations are managed by the hypervisor, a backup application just sends a request to create or remove a snapshot.

However, in some environments, VMs can lose connection during snapshot removal, I suggest checking through this KB to get more information about this issue. I don't think that it's possible to estimate the likelihood to face this problem in every specific environment, the best way to go is to test it, probably out of production hours.

Also, you can consider direct backups from storage snapshots if you're using a storage system from this list. In this scenario, VMware snapshot exists just for a very short period of time, thus the risk of connection issues during snapshot removal can be significantly mitigated.

Thanks!
dloseke
Service Provider
Posts: 66
Liked: 29 times
Joined: Jul 13, 2018 3:33 pm
Full Name: Derek M. Loseke
Location: Omaha, NE, US
Contact:

Re: VM backup with large VMDK files

Post by dloseke »

Honestly....this will probably have less to do with the size of the CM than the IO change rate and how large of deltas are created during the backup process. This will affect the commit time for the snapshot as well as high IO can cause a longer stun time which could be noticable. That said, I have a client with a 120ish TB file server and I dont have much issue with backups though I do have the disks split into a couple of jobs because they have to go to different repositories due to disk sizing. I actually have more issues with health checks and defrag of the backup files than the actual backups.
Derek M. Loseke, Senior Systems Engineer | Veeam Legend 2022-2024 | VMSP/VMTSP | VCP6-DCV | VSP/VTSP | CCNA | https://technotesanddadjokes.com | @dloseke
Post Reply

Who is online

Users browsing this forum: Spex and 80 guests