Citrix and NVIDIA recently introduced two new vGPU types, the K120Q and K220Q already pro-active users have started trialling them. The entry level vGPU for scalability is the K100, however this primarily designed for uses like WebGL and Windows Aero with a 256Mb frame buffer and can show odd artifacts on a benchmark like Unigine Heaven that needs 512Mb of frame buffer. 512Mb of framebuffer is just what the K120Q and K220Q use, offering the same user density as the K100/K200.
NVIDIA have just launched some super forums to support NVIDIA GRID technology users that you can find here: https://gridforums.nvidia.com/, the perfect place to find out more about vGPU for VMware and Citrix platforms! The sharp-eyed amongst you will have noticed users such as Tobias Kreidl, system administrator at Northern Arizona University and virtualisation community guru has started publishing his initial results on the K120Q, he is investigating whether there is a significant reduction in performance with applications like Unigine Heaven is you choose the higher density of the K120Q versus a higher powered K140Q vGPU.
The NVIDIA forums are shaping up to be a great resource, you can read Tobias’ post and user comments or ask questions on the thread, here, in summary:
“The difference is that the K140Q setup allocates 1 GB of GPU memory and you can have a maximum of 16 vGPUs per NVIDIA GRID K1 card. The brand new K120Q format is limited to the same maximum frame size, but allocates just 512 MB of RAM and hence allows up to 32 vGPUs per board. By contrast the K100 also supported 32 vGPUs, but only with 256 MB of memory.
This demonstrates a minimum penalty for using the smaller configuration which even with large formats isn’t that much less, while essentially no difference with either smaller frames or the somewhat less demanding DirectX 11 format. In other words, this density supports as many users as the weaker K100 and sacrifices little performance compared with the K140Q. This is a nice and somewhat surprising result. “ says Tobias.
Currently available vGPU specifications