Nvidia Introduces World's First Virtualized GPU Technology

By - Source: Toms IT Pro

Five years in the making, Nvidia's virtualized the GPU to bring you graphical power on your VDI.

Five years in the making, Nvidia's virtualized the GPU to bring you graphical power on your VDI.

Cloud was a big topic at Nvidia's GPU Technology Conference, which wasn't unexpected given the focus of GPUs in servers and enterprise applications.

At a keynote speech, Nvidia CEO Jen-Hsun Huang announced the fruit of the company's five-year-long effort in creating the world's first virtualized GPU, dubbed the VGX platform, which enables IT departments to deliver a virtualized desktop with the graphics and GPU computing performance of a PC or workstation to employees using any connected device.

Nvidia CEO Jen-Hsun HuangNvidia CEO Jen-Hsun Huang

While virtualized CPU computing is commonplace today, Nvidia VGX enables workers for the first time to access a GPU-accelerated desktop similar to a traditional local PC, which is hugely advantageous for 3D design and simulation tools – applications which had previously been too intensive for a virtualized desktop.

Nvidia's Vision of a GPU-Accelerated DesktopNvidia's Vision of a GPU-Accelerated Desktop

Nvidia also pitched the VGX platform as a solution for enterprise IT departments to address the complex challenges of "BYOD" – or bring your own device to work. It delivers a remote desktop to these devices, providing users the same access they have on their desktop terminal. At the same time, it helps reduce overall IT spend, improve data security and minimize data center complexity.

The BYOD Problem for ITThe BYOD Problem for IT

The Nvidia VGX hardware is designed for hosting large numbers of users in an energy-efficient way. The first of such VGX board packs four GPUs, 16 GB of memory, 4 GB of frame buffer, and fits into the industry-standard PCI Express interface in servers. Perhaps even better is that the VGX boards are designed to be passively cooled.

On the software side is the GPU Hypervisor, a software layer that integrates into commercial hypervisors, such as the Citrix XenServer, enabling virtualization of the GPU.

The VGX package also features Nvidia User Selectable Machines (USMs). This manageability option allows enterprises to configure the graphics capabilities delivered to individual users in the network based on their demands. Up to 100 users can be served from a single server powered by one VGX board, which Nvidia says will improve user density on a single server compared with traditional virtual desktop infrastructure (VDI) solutions.

Read more about Nvidia's big week at our liveblog of the GTC keynote.

Take your big ideas off the back burner with Converged Infrastructure

Comments