29
Dedicated GPU
With GPU passthrough, you can create a VM with a dedicated GPU. This configuration
provides user experience comparable to using a fat client with a high-end graphics
card. However, assigning a GPU core to a single VM, either a hosted shared desktop
(SBC) or a hosted private desktop (VDI), limits scalability.
Shared GPU
Grid technology lets multiple virtual desktops share a GPU, while offering the same
user experience as native GPUs. This sharing is commonly known as a vGPU and is a
function of the Grid software and hypervisor. An NVIDIA Grid M10 card, for example,
has four physical GPU cores that can host up to sixteen users per core, resulting in
64 users, each with a vGPU-enabled desktop, per M10 card.
The GPU processes VM graphics commands directly, which means that users get
high-end graphics without a performance penalty due to hypervisor interference.
vGPU is more scalable than passthrough, as we assign vGPU profiles to our users
and thus get more users on the same card.
vGPU profiles deliver dedicated graphics memory through the vGPU Manager,
which assigns the configured memory for each desktop. A vSphere installation
bundle (VIB) installs the vGPU Manager on the hypervisor. With AHV, an RPM
package manager performs this task. Each VDI instance has preset resources
based on the applications' needs.
Grid Licensing
Something unique to NVIDIA is that to utilize the vGPU functionality, licensing is
required. There are different levels of licensing for virtual apps that are used for RDSH
based solutions. Power user and designer for most VDI use cases covers high-end
applications such as Adobe applications, engineering, and CAD applications. The Grid
licensing is available in named or concurrent user options.