Nvidia and VMware officials are looking to bring high-end graphics capabilities to hosted desktop environments, opening up the computing model for engineers and others who use three-dimensional and other graphics-rich applications that are delivered via the cloud.
At the same time, the two companies are working on solutions that will enable VMware customers to use Nvidia’s Grid vGPU (virtual GPU) with VMware virtual machines, enabling them to share GPU acceleration capabilities between multiple virtual desktops without impacting the quality of the graphics, the companies said.
The remote desktop news was among a series of announcements made this week by Nvidia CEO Jen-Hsun Huang at the company’s Graphics Technology Conference 2014.
The desktop-as-a-service (DaaS) model gives businesses a range of benefits, from greater scalability to improved security to increased employee productivity. DaaS has enabled businesses to give remote workers easy access to their PC images from wherever they are, while keeping the data safe by having it centrally stored in the data center. However, those desktops and applications were delivered with only CPU capabilities, leaving out the graphics components, according to Justin Boitano, general manager of cloud and virtualization solutions.
GPUs are important components of most computing systems, from workstations and notebook PCs to tablets and smartphones, Boitano said in a guest post on the VMware blog.
As a result, many users—particularly those running graphics-intensive applications—preferred more powerful and reliable physical systems,” he wrote.
That will change, thanks to the partnership between Nvidia and VMware, Boitano wrote. Now, Nvidia’s Grid GPUs, built on the company’s Kepler architecture, are available on VMware’s Horizon DaaS platform, enabling mobile access and remote collaboration for workers and teams in multiple locations to work with graphics-intensive 3D applications.
“VMware Horizon DaaS users can now, for the first time, experience even the most demanding graphics-intensive applications virtually, through any connected device, while seamlessly shifting from private to public clouds,” Boitano wrote.
The access to GPU capabilities in remote desktop environments will mean that such applications as Adobe’s Photoshop and Autodesk auto-CAD will be able to run better in DaaS scenarios, according to Patrick Moorhead, principal analyst at Moor Insights and Strategies.
“These benefits are obvious, but the limitations relate to the fact that traditional desktop virtualization simulates the functioning of the GPU on the CPU,” Moorhead wrote in a column in Forbes. “As you would expect, any remote app that uses only the CPU will be very slow, particularly when it comes to watching videos and surfing an advanced Web page is painfully slow.”
He said the combination of Nvidia’s Grid GPU technology with VMware’s DaaS solution and its ESXi hypervisor technology “means that enterprises can now support literally any app their user base wants, with the highest level of graphics performance, at the highest levels of efficiency.”
The capability will be a boon for VMware, its remote desktop users and for Nvidia, which is partnering with the company that holds 70 percent to 80 percent of the worldwide hypervisor market, according to Moorhead.
Nvidia’s Grid GPUs are available now to service providers and users of VMware’s Horizon DaaS platform. VMware officials said an early access version of the Grid GPUs working with VMware virtual machines will be available to certain customers later this year.