CUDA: Week in Review
Monday, August 23, 2010, Issue #33 - Newsletter Home
Welcome to "CUDA: Week in Review," an online news summary for the worldwide CUDA and GPU Computing community.
– Contact us at:
– Follow us on Twitter:
– See previous issues:
– Register for GPU Technology Conference (Sept. 20-23):
   (code: GMCUDANEWS10)
GTC 2010 Update
We have an incredible line-up of technical content for the GPU Technology Conference (GTC 2010), covering many dimensions of the GPU ecosystem. Here's a sampling of newly-confirmed sessions:

      – GPU-Enabled Biomedical Imaging
         Homer Pien, Mass General Hospital and Harvard Medical School

      – TSUBAME 2.0
         Satoshi Matsuoka, Tokyo Institute of Technology

      – Domain Specific Languages
         Hanspeter Pfister, Harvard University

      – Evolution of GPUs for General Purpose Computing
         Ian Buck, Software Director, GPU Computing, NVIDIA

For more info, check out our recent GTC blog post. We look forward to seeing you in San Jose!
Our World: 2.1 Billion Years Ago
An international team has discovered multi-cellular organisms in Gabon, Africa that are 2.1 billion years old -- with assistance from the GPU. This discovery was recently highlighted in Nature. To learn more, we contacted Arnaud Mazurier of ERM in Poitiers, France and Francois Curnier, CEO of Digisens.
NVIDIA: Arnaud, why is this discovery a breakthrough?
Arnaud: The first chapter of the history book on life now needs to be rewritten! This discovery is quite astonishing because the fossils (which are 1-12 centimeters long and look a bit like cookies) reveal that large organisms were growing in a coordinated manner 2.1 billion years ago, rather than ˜600 million years ago as previously thought.

NVIDIA: What are the implications from a scientific perspective?
Arnaud: This discovery can potentially provide scientists with new insight into the planet Earth during that time, including atmospheric and ocean conditions. It causes us to rethink our conclusions about how multi-cellular life first appeared and evolved.

NVIDIA: How did GPU computing play a role?
Arnaud: The use of GPUs (Quadro FX5600, Tesla C870) allowed us to reconstruct data more quickly (in this case, 6-10 minutes with GPUs versus 12-15 hours in the past with CPUs). This gave us precious time to examine more specimens and refine the final virtual reconstruction.

NVIDIA: Who are the key players on the research team?
Arnaud: This is a multidisciplinary project coordinated by Abderrazzak El Albani from the University of Poitiers. In all, 16 institutions and 21 researchers are involved. ERM was in charge of acquisition and virtual 3D reconstructions, using DigiCT software from Digisens.

NVIDIA: Francois, tell us about Digisens.
Francois: We create reconstruction software that transforms 2D projections into 3D volumes. Most people know about this technology through 'computed tomography' (CT) scans. An innovative research team at the University of Poitiers was one of our first customers. They were very open-minded about GPU computing back in 2007.

NVIDIA: How does CUDA help you?
Francois: Because we are pushing the capabilities of the GPU to the max, we rely on the CUDA C environment. Moreover, we appreciate being part of the CUDA ecosystem. Our company has a global GPU approach to technology, whether we are introducing algorithms for reducing X-ray dosages or delivering better images for nanotechnology researchers.

For info on Digisens, see: For info on ERM, see:
Georgia Institute of Technology Named CUDA Center of Excellence
Georgia Institute of Technology (Georgia Tech) has been named a CUDA Center of Excellence. Jeffrey Vetter of Georgia Tech and Oak Ridge National Laboratory will serve as principal investigator. Prof. Vetter commented: "Georgia Tech has a long history of education and research that depends heavily on the parallel processing capabilities that NVIDIA has introduced with its CUDA architecture." See:
CUDA Survey
CUDA Users: Please take a few minutes to tell us how you are using CUDA-capable GPUs and how we can better meet your needs.
MATLAB Benchmarks
John Melonakos of Accelereyes recently posted a blog entry comparing Tesla C2050 versus Tesla C1060 on MATLAB apps. He writes: "Double-precision examples on the Fermi-based board outperformed the older board by 50% in every case and better than 2X in many cases."
Post-GTC Workshop from SagivTech
SagivTech will hold a one-day workshop on Friday, September 24 at NVIDIA (following GTC) with a focus on CUDA optimization. Discount for GTC attendees. See:
New Parallel Nsight Webinars
–  Overview of Parallel Nsight 1.0 for Microsoft Visual Studio
    Aug. 23, 8:00 pm –

–  Debugging Massively Parallel Apps with Parallel Nsight 1.0 / Microsoft
    Visual Studio
    Aug. 25, 9:30 am –

–  Analyzing and Optimizing Massively Parallel Apps with Parallel Nsight
    1.0 / Microsoft Visual Studio
    Sept. 1, 9:30 am –

–  Debugging and Analyzing Graphics Apps with NVIDIA Parallel Nsight
    1.0 / Microsoft Visual Studio
    Sept. 8, 9:30 am –
GPU Computing Webinars from NVIDIA
– For info on all upcoming webinars, see:
Training from SagivTech
– CUDA course: Sept. 27-29, San Francisco (following GTC 2010):
– GPU Computing@30,000 feet:
Training from Acceleware
– Sept. 13-17, Calgary:
Training from EMPhotonics
– On-site standard and customized training programs:
CUDA Certification
– New certification program for GPU computing developers:
CUDA and Academia
– Over 350 universities are teaching CUDA and GPU Computing courses around the world.
– The CUDA Center of Excellence Program recognizes universities expanding the frontier
   of parallel computing.
– The CUDA Research Center Program recognizes institutions performing leading-edge
– The CUDA Teaching Center Program recognizes universities providing education and
   hands-on instruction.
– The Academic Partnership Program provides support to researchers using GPUs to
   solve the world´s most challenging problems.
– Learn more about NVIDIA´s Research and University activities at
– Download CUDA 3.1 Toolkit:
– OpenCL v1.1 pre-release drivers and SDK code samples are available to GPU Computing
   registered developers. Log in or apply for an account to download.
August 2010

Symposium on Chemical Computations on GPGPUs
Aug. 22-26, Boston

Unconventional High Performance Computing 2010 (UCHPC 2010)
Aug. 31-Sept. 1, Italy

September 2010

GPU Technology Conference (GTC) 2010
Sept. 20-23, San Jose, Calif. (register today, space is limited)

MATLAB Conference 2010
Sept. 30, Wembley Stadium, London


Supercomputing 2010
Nov. 13-19, New Orleans, LA

IEEE International Parallel & Distributed Processing Symposium
May 16-20, 2011, Anchorage, AL

(To list an event, email:

CUDA Articles in Dr. Dobb's
– Supercomputing for the Masses, Part 19:
– Supercomputing for the Masses, Part 18:
– Supercomputing for the Masses, Part 17:
– Supercomputing for the Masses, Part 16:
CUDA Books
– CUDA By Example by J. Sanders, E. Kandrot:
– Programming Massively Parallel Processors by D. Kirk, W. Hwu:
– See additional books here:
CUDA Documentation
– Download developer guides and documentation:
– Read previous issues of CUDA: Week in Review:
– Follow CUDA & GPU Computing on Twitter:
– Network with other developers:
– Stayed tuned to GPGPU news and events:
– Learn more about CUDA on CUDA Zone:
– Read Kudos for CUDA:
About CUDA
CUDA is NVIDIA’s parallel computing hardware architecture. NVIDIA provides a complete toolkit for programming on the CUDA architecture, supporting standard computing languages such as C, C++ and Fortran as well as APIs such as OpenCL and DirectCompute.

Send comments and suggestions to:

Click here to opt in specifically to CUDA: Week in Review.

Copyright © 2010 NVIDIA Corporation. All rights reserved.
2701 San Tomas Expressway, Santa Clara, CA 95050.