NVIDIA CEO Updates NVIDIA’s Roadmap


Our CEO, Jen-Hsun Huang, sketched out today a future where mobile, desktop and supercomputer technologies intersect in powerful and surprising ways.

Speaking at the GPU Technology Conference, Huang described how GPUs are already finding their way into applications that were undreamed of a decade ago.

NVIDIA’s Kepler GPU architecture – introduced last year — is not only a runaway hit with gamers. It underpins a new generation of hyper-efficient supercomputers, including Titan, the world’s fastest system.

Next up: new GPUs and mobile processors that promise to put the efficiency and speed of our massively-parallel GPU architectures into an ever-broader array of devices.

In GPUs: 

  • Maxwell will offer unified virtual memory, giving CPUs access to the speedy memory built into GPUs, and vice versa.
  • In Volta, due after Maxwell, memory modules will be piled atop one another and placed on the same silicon substrate as the GPU core itself – a radical idea called ‘stacked DRAM’ – giving these new GPUs access to up to one terabyte per second of bandwidth. That’s enough to move the equivalent of a full Blu-Ray disc worth of data through a chip in just 1/50th of a second. “Unbelievable stuff,” Huang said.


In mobile processors:

  • Logan will pair ARM-based mobile processor cores with our powerful Kepler GPUs, putting technologies now found in high-performance PCs and workstations – such as PhysX, CUDA 5, and Open GL 4.3 – into mobile devices.
  • Parker will join new 64-bit ARM compatible CPU cores with our next-generation Maxwell GPU architecture – combining the ability to gulp down big chunks of data, like a server, with Maxwell’s ability to mix and match memory resources with CPUs.


Note: an earlier version of this post described Volta’s use of ‘stacked DRAM’ incorrectly. 



How Harley-Davidson Uses GPUs and 3D Modeling to Cut Months Off Its Design Cycle


Harley-Davidson is using tools from the movie and gaming industries to shave months off the design and development of its motorcycles.

Matthew Gueller, a senior industrial designer for the company, told the GPU Technology Conference about how his eight-person design team is able to crank out new design concepts in hours, even minutes.

The veteran industrial designer said Harley has gradually moved away from traditional styling tools, which entailed sketches and clay prototypes requiring painstakingly slow iterations. By adopting 3D modeling tools and 3D printing, the design team can create new prototype components in less than 24 hours.

Harley uses a suite of Autodesk applications – including 3D Studio Max, Maya, Mudbox, and Alias – running on workstations powered by NVIDIA’s Quadro and Tesla GPUs

These are “not traditional tools for Harley Davidson – they are from the gaming and entertainment industry,” Gueller said. “We love it, it’s awesome.”

He said the modeling done with these programs is a “paradigm shift that’s a game changer.”

The tools have helped his styling team develop a component model in 5-10 minutes, and get a new prototype from the 3D printing team by the next morning. Four or five days were required using traditional methods.

Taken together, programs running on GPU-based systems “shave months off the development of new motorcycles,” he said.

“It used to take four people on our team a few months to develop a new bike design. Now, two people can design a full motorcycle in two weeks,” he concluded.

Harley-Davidson's TK

Harley-Davidson’s Matthew Gueller talks about modern motorcycle design.

AMD To Report Fiscal First Quarter Results On April 18, 2013

AMD (NYSE: AMD) today announced that it will webcast its quarterly earnings conference call on Thursday, April 18, 2013 at 5:00 p.m. EDT / 2:00 p.m. PDT to discuss the results of its fiscal first quarter ending March 30, 2013. All interested parties will have the opportunity to listen to the real-time audio webcast of the teleconference over the…

Swiss National Supercomputing Center Plans to Scale Application Performance Heights With NVIDIA GPUs

Swiss National Supercomputing Center Piz Daint supercomputer

Ask any weatherman: Predicting the weather is not easy. Throw in Switzerland’s many diverse microclimates and topographical features and the challenge is even greater.

The Swiss National Supercomputing Center (CSCS), one of Europe’s top institutions for computational research, plans to change that by applying a petaflop of GPU-accelerated processing power to the problem.

CSCS is building a new Cray XC30 supercomputer called “Piz Daint,” named after a mountain in the Swiss Alps. Piz Daint will be extended with GPU accelerators to dramatically expand the breadth and depth of the center’s research and discovery in climate and weather modeling, as well as a host of other fields, such as astrophysics, materials science and life science.

In particular, CSCS is working with MeteoSwiss, Switzerland’s national weather service, to reach the holy grail of meteorology: predicting local and national weather patterns days or even weeks ahead of time with the highest degree of accuracy.

With NVIDIA Tesla K20X GPU accelerators, Piz Daint will have more than 1 petaflops of performance – that’s 1,000 trillion floating point operations per second – which is expected to make it the fastest GPU accelerator-based scientific supercomputer in Europe when it becomes operational in early 2014.

Based on the NVIDIA Kepler architecture – the world’s fastest and most energy-efficient high performance computing architecture – the Tesla GPUs will dramatically accelerate performance at an affordable cost. This is key as running complex, compute-intensive simulations of large-scale environmental phenomenon accurately takes massive computing resources.

This is an ideal task for a GPU-powered supercomputer, but beyond the abilities of typical CPU-based systems, and certainly impossible for CPUs to do quickly.

“Piz Daint will help advance our research into alpine climate and weather patterns by leaps and bounds,” said Thomas Schulthess, director of CSCS. “With GPU acceleration, researchers can run many more sophisticated, ultra-high-resolution models, giving us an unprecedented level of visibility and understanding into how these systems work.”