Brian published an article in which he mentions a few interesting things. If you read my article about the grid products you have to find your way what is what, so here’s the manual on how to do that;
- vSGA = Shared GPU
- vDGA = GPU pass-through,
- vGPU = Virtual GPU
I like the acronyms Brian uses more than the nVidia’s because there is less chance on confusion for the different technologies. It’s almost the whole Citrix product name confusion all over again, so i hope he and i made it a bit more clear for you.
He mentions some other interesting news;
- Horizon view 5.3 supports GPU pass-through (vDGA)
- ESXi isn’t adding vGPU support until 2015.
- He (and nVidia) also mentions vDGA is a GPU (core) per user, i think that should be per VM, so that could be an RDSH/XenApp server too and that would mean more than 1 user. Read my article about the grid products for more details.
By the way, I (still) haven’t heard any news about vGPU Hyper-v support. Seems this vGPU stuff is slooow in adaption, although my use cases focus on vDGA so we’re ok for now.