topiauf.blogg.se - Veeam 9.5 backup copy best practices

If you have a multiple exchange servers in a DAG I wouldn't backup all of them all the time, or simultaneously. This depends on your infrastructure design. Having identical VMs in the same job was quite a bit of space one could save if used with traditional job chains, but not so much when using per-VM backup files.

So if you're using per-VM backup files and have 300 VMs in your job, the job creates 300 chains, thus lowering the savings. When using per-VM backup files the savings from VEEAM will be lower because the data savings occur 'per-job-chain'. Of course this is mitigated by the SOBR, but it's still an inconvenience. Huge jobs will always need to move at once, while multiple smaller jobs could move one at a time to another repo.

I used bigger jobs in the past, but they're aren't good to manage at some point.

If you're using per-VM backup files there's a recommendation to use no more than 300 VMs.

I'm unaware on how resource intensive this is so I figure with most things it's a matter of finding a balance between frequency and performance impact on the repository server.

How frequently should corruption guard be scheduled to run?.

Can anyone help me decide if I should enable or disable them? I hear that synthetic fulls are better with ReFS due to fast clone, however I've read a few posts on this subreddit that mention that synthetic fulls aren't really needed and to leave them disabled. I am using ReFS on my repository and incremental backup mode for my jobs. With other backup products I have been careful to stagger jobs based on their estimated duration so that the backup server is essentially always busy but not overloaded, is this not a concern with Veeam? Does this mean I shouldn't be concerned with schedule overlap between backup jobs? I hear about Veeam's powerful automatic resource scheduler.

I am also looking for suggestions on job scheduling.

What should determine whether a VM or group of VM's gets placed into a new backup job?Īssuming the number of restore points are equal for all VM's what factors would make me consider a single backup job with many VM's vs multiple backup jobs with less VM's per job?įor example is there an advantage to splitting SQL servers into a different backup job than Domain Controllers?.

We will use backup copy and GFS to satisfy 3-2-1. The majority of our VM's to backup are Windows file and SQL servers that I plan to retain on our primary repository for anywhere from 10 to 30 days. I am using a Hyper-V off-host proxy that accesses source VM storage directly from the SAN using hardware VSS snapshots. I have a few remaining questions about the best way to configure backup jobs. I'm new to Veeam but have been using their excellent help center site to get started with my veeam B&R enterprise deployment.