510.226.7366 | sales@exxactcorp.com |

Workstation Graphics and HPC Solutions  

Exxact Corp - Professional Graphics, Consumer Graphics, GPU Computing, Visual Computing, Other Products

NVIDIA GRID - GPU Virtualization for Virtual Desktops and On-Demand Application Delivery

Imagine if you could access to your GPU-accelerated applications anywhere on any device, even those require intensive graphics power. NVIDIA GIRD makes this possible. With NVIDIA GRID, a virtualized GPU designed specifically for virtualized server environment, data center managers can bring true PC graphics-rich experiences to more users for the first time.
The NVIDIA GRID GPUs will be hosted in enterprise data centers and allow users to run virtual desktops or virtual applications on multiple devices connected to the internet and across multiple operating systems, including PCs, notebooks, tablets and even smartphones. Users can utilize their online-connected devices to enjoy the GPU power remotely.


NVIDIA GRID for Virtual Desktop Infrastructure (VDI)

NVIDIA GRID for On-Demand Application Delivery

NVIDIA's Kepler™-based GRID K1 and K2 boards are specifically designed to enable rich graphics in virtualized environments.
NVIDIA Driver:
The NVIDIA driver is needed for full graphics functionality and performance. This ensures application and graphics API compatibilities in the operating system.
GRID Enabled Virtual Desktop:
A GRID enabled virtual desktop uses the GRID technology to quickly capture and compress the frames to be sent over the network, and also reduce the latency of the overall remote experience to be more like a physical desktop experience.

Blade server support:
With GRID 2.0, run GRID-enabled virtual destops on blade servers -- not simply rack servers.

Linux support:
NVIDIA GRID 2.0 enables enterprises in industries that rely on Linux applications and workflows to take advantage of graphics-accelerated virtualization.
Hypervisor is virtualization host platform that incorporates the ability to access GPUs directly from the virtual machine.
Citrix produces the XenServer hypervisor for server and desktop virtualization. XenServer 6 is a GRID enabled hypervisor with pass-through GPU today, and will be the first to market with the GRID vGPU. Hyper-V is Microsoft’s hypervisor that can run on its own as a native hypervisor, or from within Windows Server 2008 R2 or Server 2012. Hyper-V is not a GRID enabled hypervisor, but it can take advantage of GRID hardware through Microsoft’s software virtualization solution, RemoteFX. ESXi, also known as vSphere, is the hypervisor solution from VMware. ESXi 5.1 is a GRID enabled hypervisor with pass-through GPU today.

GPU Virtualization Technology

NVIDIA GRID for Virtual Desktop Infrastructure (VDI)
(1) Pass-through GPU (One-User per GPU):
Figure 1. Pass-through GPU
Also known as “vDGA” (VMware) and “GPU pass-through” (Citrix).
This method allows discrete GPU devices to be directly mapped to a virtual machine for dedicated 1:1 use by the VM (see figure 1). The virtual machine has full and direct access to GPU to utilize full workstation graphics and GPU compute performance.
The pass-through GPU model is ideal for graphic-intense 3D applications and GPU Compute users so that GPU device is directly owned and managed by the VM operating systems just as in a desktop workstation.
GRID Enabled Hypervisors using this technology include:
• Citrix XenServer 6.0 and newer
• VMware vSphere 5.1 and newer
> Exxact GRID Solutions for Pass-through GPU
(2) vGPU (Multi-User per GPU):
Figure 2. vGPU
vGPU is the Citrix implementation of this technology allowing single GPU mapped to multiple users. This offers the benefit of a single GPU scaling to many virtualized GPUs, as well as provides the performance of the pass-through models (see figure 2.)
The GRID GPU is shared between multiple VMs, and each VM has direct access to the dedicated GPU via the NVIDIA GRID vGPU Manager. The NVIDIA GRID vGPU Manager within the host hypervisor manages the VMs to GPU channels, guaranteeing each VM has a dedicated amount of vRAM and direct access to the GPU. IT managers can assign different GPU power usages to multiple users depending on their workload needs.
GRID Enabled Hypervisors using this technology include:
• Citrix XenServer with NVIDIA GRID cards and GRID vGPU technology
> Exxact GRID Solutions for vGPU
NVIDIA GRID for On-Demand Application Delivery
In this model, 3D applications are installed on the host Windows server and published via hosted shared desktops or as a hosted published application. The Installed applications can utilize GPU power for 3D rendering from the server equipped with NVIDIA GRID GPUs.
Figure 3. On-Demand Application Delivery
Supporting up to 300+ users on a single server, this model offers a significant cost saving and the ability for centralizing and managing all applications in a data center.
GRID Enabled Hypervisors using this technology include:
• Citrix XenApp w/DirectX Hosted Apps and Desktops
• Citrix XenApp w/OpenGL Hosted Apps and Desktops (Available as of XenDesktop 5.6 FP1 and 7)