Backup Infrastructure Design Help

[timstumbo]

I was hoping some people could share on how they have their VMWARE backup infrastructure setup. I've been having a lot of issues lately with the backup server locking due to high CPU ussage so I'm wondering if I have things setup the most efficient way possible.

I've read the http://helpcenter.veeam.com/backup/80/v ... ments.html page but I'm looking for opinions from personal experience.

I have 7 backup jobs totaling 26 VM's. I have copy jobs for each of these backups that copies the data offsite.

I have a single VM setup as the backup server and the proxy. The VM has 1 virtual socket with 2 cores and 12GB of memory. Is this adequate? Also should I look at setting up a secondary server as the proxy?

I use an onsite and offsite NAS for storage connected via iSCSI.

Thanks

[veremin]

So, you have 14 jobs in total, how many concurrent jobs are typically running? We recommend having one core per parallel job. Thanks.

[timstumbo]

Well currently I have the proxy and the backup repositories to only allow a single job to run at a time. However, if I could get things running better it would be nice to have the copy jobs run the same time as the backup jobs to cut down the backup window times.

[veremin]

What compression settings are specified? Is it high by any chance? Thanks.

[timstumbo]

Yes, under the job settings the compression level is set to high, and under strorage optimization I have LAN Target selected since were backing up to a NAS.

[dellock6]

Probably lowering the compression level to optimal would increase processing speed, while at the same time the backup file size will not be impacted that much.
As a second check, maybe trying also to increase the cores to 4 and increase concurrent jobs to 2, while always having a look at bottleneck stats to see which component is more impacted by the settings.

[timstumbo]

Thanks, I will make the changes and see how the jobs run tonight.

[veremin]

Let us know how it goes. Generally speaking, lowering compression level to optimal should decrease the load put on a backup server at the cost of small backup file's growth. Thanks.

[timstumbo]

I modified all the settings and changed the hardware on the server to 4 cores. After adjusting the hardware the server seemed to be even more unstable. I know that this is always a risk when making those adjustments after setting up a VM. Regardless, the server locked up again last night during a copy job. I reviewed the system logs in event viewer but they didn't seem to be very helpful.

I'm dreading asking this question but I'm going to throw it out there anyway. Could this have anything to do with running SQL Express instead of Standard?

[timstumbo]

Server is still locking up. One thing I've noticed is that it doesn't seem to do it when I have all my copy jobs disabled. If I'm just running the primary backup jobs so only one process is running on the server at a time things seem to be functioning ok.

Also, any idea if using SQL Express could be contributing to this problem?

[dellock6]

SQL Express is considered a bottleneck only if you have large environments, like above 500 VMs or such... far beyond your size.
At this point, I can suggest you Tim to open a support ticket and let support investigate logs.

[foggy]

Job bottleneck analysis should also help in identifying the issue and I also wonder how your repositories are added into Veeam B&R console.