Hi Simon,

Sorry for the late response.  A while ago I posted a message with my
setup and some speed results:

http://mailman.eng.auburn.edu/pipermail/veritas-bu/2011-March/109633.html

We have hundreds of VMs and we indeed limit the amount of jobs per
datastore (thru the global option available for this).  Our tape library
has 4 drives (LTO4) and we're using NBU 7.1.

The main criteria we use to separate the policies is whether or not we
use vRay on them (that's the Mapped Full VM option).   We had some
issues with some Linux VMs with ext3 filesystem where the "Mapped Full
VM option" was slowing down the backups.  I've been told this issue has
been addressed on the latest version of Netbackup.

Feel free to ask any other specific question.

Best regards,
Jorge




On 02/27/2012 02:25 PM, Simon Weaver wrote:
> All
> Having spent many months working on this, and having it setup to my
> liking, I wanted to get a feel if any other NBU Admin's are doing
> anything similar to what I am proposing.
>  
> 1) Policy has many clients, all local drives, using DNS Name and
> allowing for Full and Incr Backups
> 2) Limit the amount of jobs that can run concurrently. At the moment I
> have it set for 4
> 3) Backup directly to tape
>  
> I have several hundred VM's and I am concerned that I may not be able to
> backup them all up quick enough over SAN. I was thinking of adding more
> Policies and splitting the clients across them, maybe on a Datastore
> basis.
>  
> Can I have some thoughts or input on maybe the best way to perform this
> or what works well for you?
>  
> Thanks, Simon
> 
> 
> 
> 
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu