Nvidia has rolled out a slew of new products, from a high-speed GPU connect that may pave the way for exascale computing to a new product design that could replace physical prototypes with interactive, photorealistic digital models and more.
We caught up with Charles King, principal analyst at Pund-IT, to get his take on Nvidia’s market moves. He told us the company is going after a much broader swath of markets than it has ever pursued in the past.
“You could say that the announcements provide kind of a seed of evidence of the company’s broad market strategy and a very clear snapshot of a company that will not limit its ambitions to traditional markets,” King said. “Nvidia wants to play in areas that might be complementary to, but are also far beyond its traditional market in graphics cards and gaming.”
With that in mind, let’s take a closer look at the announcements.
The Power of NVLink
First, Nvidia announced plans to integrate a high-speed interconnect, dubbed NVLink, into its future GPUs. The company claims its innovation would allow GPUs and CPUs to share five to 12 times faster than currently possibly and do away with bottlenecks. Ultimately, the tech could set the stage for a new generation of exascale supercomputers that are 50 to 100 times faster than today's most powerful systems.
Nvidia is adding the NVLink technology into its Pascal GPU architecture, which is scheduled to launch in 2016. The new interconnect was co-developed with IBM, which is incorporating it into future versions of its POWER CPUs.
"NVLink enables fast data exchange between CPU and GPU, thereby improving data throughput through the computing system and overcoming a key bottleneck for accelerated computing today," said Bradley McCredie, vice president and IBM Fellow at IBM.
"NVLink makes it easier for developers to modify high-performance and data analytics applications to take advantage of accelerated CPU-GPU systems. We think this technology represents another significant contribution to our OpenPOWER ecosystem," he said.
Nixing Physical Prototypes?
Next, Nvidia took the lid off a GPU rendering appliance that it said “dramatically accelerates” ray tracing. The idea is to make it possible for professional designers to interact with computer models of such high visual fidelity that it can essentially replace the costly and cumbersome process of building physical prototypes.
It’s called the Iray Visual Computing Appliance (VCA), and it combines hardware and software to accelerate the work of Nvidia Iray -- a photorealistic renderer integrated into design tools like Dassault Systèmes' CATIA and Autodesk's 3ds Max. Iray VCA systems will be available starting in summer through a global VAR network of certified system integrators. Pricing is $50,000 in North America. Honda is already using it.
"For our styling design requirements, we developed specialized tools that run alongside our RTT global standard platform," said Daisuke Ide, system engineer at Honda Research and Development. "Our TOPS tool, which uses Nvidia Iray on our Nvidia GPU cluster, enables us to evaluate our original design data as if it were real. This allows us to explore more designs so we can create better designs faster and more affordably."
Nvidia is also partnering with VMware to make Nvidia GRID technology available on the VMware Horizon DaaS (desktop-as-a-service) platform. The platform makes possible 3D graphics on virtualized desktops and applications delivered through the cloud. Nvidia said DaaS can improve productivity by making mobile access and remote collaboration for project teams or contractors in multiple locations a reality. The technology is immediately available.
"VMware Horizon DaaS Platform with Nvidia GRID is a great solution for companies with high-powered computing and visual graphics needs," said Sumit Dhawan, vice president and general manager of End-User Computing at VMware.
"It enables businesses to drive innovation via instant mobile access and collaboration across users worldwide by delivering exceptional 3D quality graphics applications from the cloud. With VMware and Nvidia working together, customers and service providers benefit from working with two industry leaders in their respective fields," he added.
Mobile Supercomputer for Embedded Systems
Finally, Nvidia believes it has paved the way for a new generation of applications that tap computer vision, image processing and real-time data processing with the Nvidia Jetson TK1 Development Kit, which leverages compute capabilities to make possible new computer-vision applications for the robotics, medical, avionics and auto industries. The Nvidia Jetson TK1 Development Kit can be pre-ordered for $192 in the U.S.
"Jetson TK1 fast tracks embedded computing into a future where machines interact and adapt to their environments in real time," said Ian Buck, vice president of Accelerated Computing at Nvidia. "This platform enables developers to fully harness computer vision in handheld devices, bringing supercomputing capabilities to low-power devices."