Sometimes I don’t hear a rumble until it becomes a roar. I’m not sure if CUDA has become a roar yet, but my ears have perked up based on a bunch of announcements I’ve received over the past few months. If CUDA hasn’t registered on your radar yet, here’s a brief summary.

CUDA, which stands for Compute Unified Device Architecture, is a C language environment developed by Nvidia Corp. (www.nvidia.com) to solve complex computational problems in a fraction of the time it usually takes using conventional methods. With CUDA, programmers can create software that taps into the many-core parallel processing power of graphics processors or GPUs.

For example, Manifold.net (www.manifold.net), which is a leading supplier of Geographic Information Services (GIS), has converted its software to the CUDA platform. With the CUDA configuration, calculations that previously took 20 minutes to complete are now done in 30 seconds, while those that took 30 to 40 seconds are now real-time. More importantly, I think, is the reaction of people like Dimitri Rotow, a product manager at Manifold.net.

“It is not an exaggeration to say that Nvidia CUDA technology could be the most revolutionary development in computing since the invention of the microprocessor,” Rotow said. “It’s fast, inexpensive, and loaded with potential. Nvidia CUDA is so important that all Manifold users should insist that the computer hardware they procure is CUDA-enabled.”

CUDA CENTER OF EXCELLENCE
Another news item that caught my eye concerned a collaboration between Nvidia and the University of Illinois at Urbana- Champaign (UIUC). Back in June, UIUC was named as the world’s first CUDA Center of Excellence. In addition to the appointment, Nvidia donated $500,000 to the university for the development of parallel computing facilities and the continuation of its research programs.

The Theoretical and Computational Biophysics Group at UIUC was one of the first research groups to leverage the parallel architecture of the GPU to accelerate its research in computational biophysics. It has successfully accelerated NAMD/ VMD, a popular parallel molecular dynamics application that analyzes large biomolecular systems.

“We’re very excited to partner with Nvidia and anticipate that together we will achieve breakthroughs in biomedicine, leading to a better understanding of disease and more effective treatments,” said Klaus Schulten, Swanlund Professor of Physics and director of the Theoretical and Computational Biophysics Group at Illinois (www.ks.uiuc.edu). “This generous gift will be a great stimulus for Illinois’ team of outstanding young programmers. It will help to extend their ranks and equip them with the necessary tools to advance computing in decades to come.”

SIGNAL INTEGRITY SIMULATIONS
Last month, Agilent Technologies announced that it is working with Nvidia to accelerate signal integrity simulations using CUDA-based GPUs. The association is expected to yield the commercial release of a GPU-enabled Advanced Design System (ADS) Transient Convolution Simulator. As a result, signal integrity designers will be able to run these simulations dramatically faster than was previously possible.

“We’re very pleased to be working with Nvidia to both speed up their design cycles today and to help our customers solve their signal integrity problems much faster in the future,” said Colin Warwick, product marketing manager with the EEsof EDA division of Agilent. “In this case, Nvidia itself is the lead customer for this new blending of technologies.”

“Using Agilent’s new CUDA-enabled tools, our engineering team was able to simulate our data path in parallel,” said Tommy Lee, vice president of System Design and Manufacturing at Nvidia. “We achieved a 14-times improvement in simulation time, sped up our NPI (new product introduction) process, and further increased our design velocity.”

CUDA is available free from Nvidia via download from its site. Millions of CUDA-capable GPUs have already been deployed as well. Nvidia also recently launched CUDA U, a new section on CUDA Zone (www.nvidia.com/cuda).

Designed for students, instructors, and developers, CUDA U hosts educational resources for the CUDA programming environment. The site contains instructional material, syllabuses and curricula, and information on schools and programs that offer CUDA instruction. CUDA U can be found at www.nvidia.com/object/cuda_education.

See Associated Figure