How Businesses Can Eliminate Separate Tiers of IT Access with Virtualized Graphics

by Victoria Rege

Call it the velvet rope of enterprise graphics.

On one side are the engineers, designers and others who require high-performance graphics. They generally use powerful workstations sporting the latest video cards.

On the other side of the rope are the vast majority of users—knowledge workers, task workers and others—who receive standard PC configurations.

Two tiers. Two deployment strategies. Two IT management approaches.

Enterprises can cut the rope by deploying shared GPU solutions in a virtual desktop infrastructure environment.

They can make graphics-rich applications available to a broad spectrum of users. They can make their deployments easier to manage. And they can better preserve the security of their data. All it takes is adopting NVIDIA GRID-powered virtualization.

GPU sharing lets companies serve different tiers of workers with a flexibility that wasn’t possible with earlier virtualization offerings.

They can provide 3D-intensive applications to engineers and designers using multiple devices, just about anywhere. At the same time, they can deploy smaller “slices” of a GPU to knowledge workers and power users, who may need a high level of graphics capability only occasionally.

Since data and applications reside securely in the data center, access can even be extended to partners outside the organization. They can view and use the data without it ever leaving the data center, boosting collaboration without compromising security.

NVIDIA’s virtual graphics processing unit, or vGPU, technology is what enables hardware sharing of graphics processing. The GPU is virtualized, while the virtual machines accessed by workers run native NVIDIA video drivers for better performance. And OpenGL support gives access to more graphic capabilities.

With GRID vGPU technology, graphics commands from virtual machines are passed directly to the GPU without any hypervisor translation—and without sacrificing server performance.

Organizations can deploy a combination of Dell servers, Citrix virtualization technology and NVIDIA GRID cards to let users experience high-fidelity graphics quality and performance for their favorite applications at a reasonable cost.

One example is Turkish Aerospace Industries (TAI), Turkey’s center of technology for the design, development, manufacturing and lifecycle support of integrated aerospace systems.

Turkish Aerospace Industries
Up and away: TAI’s virtual graphics set-up covers everything from office productivity to CAD apps.

TAI has over 1,500 engineers at its central facility, and it’s growing fast. Its IT department faced significant challenges in supporting varying hardware setups and dozens of different applications—covering everything from office productivity to heavy computer-aided design. They knew they needed to simplify before expanding further.

Replacing its previous one-to-one user-to-graphics card setup, TAI is now able to expand while optimizing IT resources. Engineers get fully optimized graphics regardless of their client hardware, encouraging productivity and innovation.

Another example is Roger Williams University, a Rhode Island institution known for its architecture program. The university deployed a virtual desktop infrastructure using Dell, Citrix and NVIDIA technology to free its architectural students and researchers from the lab environment. Now, they can learn real-world architectural practices wherever the work takes them.

The school has also eliminated bulky desktops from its architecture labs, freeing up space for classrooms, where students can work with video and graphics on their own devices. It’s now rolling out virtualization to its engineering students, with the ultimate goal of making it accessible to all students.

Learn more about how organizations are using virtualized graphics to stimulate productivity, creativity and real-time collaboration.