28
A Graphics Processing Unit (GPU) is an important part of a modern desktop experience.
Physical desktop and laptop computers have always had a GPU included, but were
not important unless you wanted to play video games or do high-end graphics-intensive
work. While these still hold true, modern operating systems and new workloads such
as machine learning (MI) and artificial intelligence (AI) are also taking advantage of
GPUs to enhance their performance.
In the VDI space, GPUs are not generally required and are typically evaluated on a use
case basis to see if the added performance is required versus balancing additional costs
versus perceived value for use cases with lower requirements. Windows 10 or watching
streaming video are examples of use cases with lower requirements that can benefit
from GPU but may not deliver enough value as compared to offering an acceptable
user experience without GPUs at a lower total cost. However, as with CPU manufacturing,
GPUs are gaining performance at a lower cost and the value derived from GPUs needs
to be part of an overall evaluation when designing an EUC environment.
NVIDIA is the best performing and most known manufacturer of graphics cards designed
for desktop virtualization. AMD's graphics cards work only in certain use cases and do
not deliver the same optimizations that NVIDIA cards offer.
NVIDIA Grid boards allow GPU virtualization, enabling multiple users to share a single
graphics card. GPU virtualization not only supports higher user densities, but also delivers
native performance while accessing a virtual desktop. NVIDIA GPUs also have an engine
for H.264 encoding that offloads processes from the CPU, which further increases user
density on your hardware. NVIDIA Grid cards typically have multiple GPUs, which
improves scaling.
GPUs