Morning all,
I'm trying to get a clarification of a statement in the VMWare "Introduction to Virtual Desktop Manager" pdf. On page 15 it states:
"VDM limits the number of provisioning and power operations that can be concurrently active for each VirtualCenter to ensure that the rate of operations is not excessive. These limits are applied across all pools and desktops for each VirtualCenter."
Now, I may be missing the obvious, but it doesn't expand on that statement. So my question is 'What are the limits'? And if I have 1000 vdis hanging off a single virtualcenter instance, how many desktops powering on, and potentially being provisioned, would it take at 9 O'Clock in the morning for users to notice a lag?
Thanks
Dave
Hi Dave,
You will have to define those limits in Advanced Pool Settings in VMD. Read page 42 of this document http://www.vmware.com/pdf/vdm21_manual.pdf
Cheers,
Hi Dave,
You will have to define those limits in Advanced Pool Settings in VMD. Read page 42 of this document http://www.vmware.com/pdf/vdm21_manual.pdf
Cheers,
Ah, got it, thanks.
Can anyone advise though on real world figures? How much of the VC resources does a vdi being provisioned and powered on take? If I was to have a dual processor dual core with 32GB of RAM managing 40 ESX servers, 9 VDM brokers and 1000 vdi's, what would sensible limits be for maximum number of concurrent provisioning operations and the maximum number of concurrent power operations be?
Hi Dave,
Download another document "VDI Server Sizing and Scaling" from here: http://www.vmware.com/support/pubs/vdi_pubs.html
Hope it helps.
Cheers,
