3
Data sheet | HP ProLiant WS460c Gen8 Graphics
Server Blade
Pass-through GPU
The HP ProLiant WS460c Gen8 Graphics Server Blade delivers the industry’s first support for
bladed graphics with up to eight GPUs per blade plus enhanced memory footprint and speeds
and full PCIe Gen3 x16 GPU support—all on the proven HP ProLiant Gen8 architecture.
This solution enables a lower cost per seat through multitenancy in a virtualized environment
while keeping a dedicated GPU per user for performance, and taking full advantage of NVIDIA
driver. This is beneficial to users who utilize high‑performance graphics applications in
multiple industries.
The HP ProLiant WS460c Gen8 Graphics Server Blade provides a local workstation for
a high‑performance PC experience to end users over the network using the one of the
industry‑standard remoting protocols such as Citrix
®
HDX 3D, VMware PCoIP, and Microsoft
®
RemoteFX. Initially, the server blade was introduced as a bare metal 1:1 solution, meaning
that a client operating system was loaded on the blade for a single user. The HP WS460c
G6 systems made a giant leap forward with its industry‑first ability to virtualize the server
blade by connecting a GPU directly to a virtual machine (first supported by XenServer 6).
This allowed multiple graphics accelerated VMs to support a range of needs from media‑rich
PCs to high‑performance 3D workstations. Now with the Gen8 version, the new graphics
offering and GPU density has taken another giant leap forward.
High-performance graphics acceleration—from media-rich
PCs to 3D accelerated graphics users
True Virtual GPU
Also known as “NVIDIA GRID vGPU” is the NVIDIA/Citrix implementation of the technology,
True Virtual GPU offers the benefit of GPU scaling like the software‑virtualized GPU
(API intercept) and provides the performance of a native NVIDIA graphics driver like the
pass‑through models (see figure 1).
This technology is currently implemented by the NVIDIA GRID K1 and K2 products. The GRID
GPU is shared between multiple VMs similar to API intercept. However, in this model each
VM has direct access to the GPU via dedicated channels managed by the NVIDIA GRID vGPU
Manager. Unlike the software virtualized GPU (API intercept) model, the NVIDIA vGPU Manager
within the host hypervisor manages the VM to GPU channels, guaranteeing that each VM has
a dedicated amount of vRAM per user and direct access to the GPU. Administrators will have
ability to assign 1 to 8 users per physical GPU depending on their workload needs.
