NVIDIA Introduces Shared Visual Computing Appliance

As CAD software becomes more and more powerful and resulting models become more complex and large in size, it requires more and more horsepower under the hood of engineering workstations to run effectively. Enter NVIDIA, the somewhat undisputed leader in the 3D graphic cards market.

You could find NVIDIA cards working quietly behind the scenes everywhere in the Partner Pavilion at this year’s SolidWorks World, being that its cards are the most popular among SolidWorks users with approximately 80% of the SolidWorks market.

The most popular of which is the Quadro Pro K2000 cards that sell for only about $400, a small price to pay for significantly increased productivity, right? Need even more power? Upgrade to the company’s K6000 for $4,500, which offers a whopping 12 GB of frame buffer memory and can make even the largest assemblies and photorealistic renderings and animations move in real time.

NVIDIA’s Quadro 6000 is the first professional-class GPU to integrate high performance computing capabilities with advanced visualization techniques, transforming modern workflows.

Touted to deliver 5X faster performance over its predecessor, the Quadro K6000 can deliver 1.3 billion triangles per second, shattering previous 3D graphics limitations. For serious data crunching for applications such as CFD, these cards deliver performance gains of 8X. Seeing these high-end cards doing their thing was quite impressive though Andrew Cresci, vertical marketing General Manager of NVIDIA, saved the best for last.

NVIDIA takes on shared, visual computing

Not happy just being the leader in 3D graphics acceleration for the design world, NVIDIA has eyes on harnessing its technology for shared computing. During my briefing with the company, Cresci gave me a sneak peak at the company’s Visual Computing Appliance (VCA). The appliance can be located anywhere–in a data center or a company’s centralized IT center–and can fed graphics to nearly any computing device, including iPads, mobile devices, low-end PCs, etc. It works by compressing graphics from the centralized server-like hub from any distance with no discernible lag, at least that I saw during my demo.

As an editor, I hate to throw around meaningless words, like “cool,” but indeed it was cool. I was able to manipulate a rather large SolidWorks assembly running on a PC in the NVIDIA booth that was being fed from the company’s headquarters in Santa Clara, 400 miles away from where we were. You can pretty quickly see the advantage of VCA for engineers and designers. Imagine being able to tap the computing power of a large server running existing software from your iPad or low-end PC.

It’s hard to not see real productivity benefits VCA offers to product developers as companies continue to try to squeeze more value out of existing resources (software and hardware). It’s not cheap; $25K+ for a floating license, but certainly a tech offering we’ll be keeping an eye on in the future. Check it out yourself on the NVIDIA site.