CUDA Week in Review Newsletter Homepage
Mon., Oct. 8, 2012, Issue #81 Newsletter Home
Welcome to CUDA: WEEK IN REVIEW, a news summary for the worldwide CUDA, GPGPU and parallel programming community.
CUDA TECH TIP: How to detect CUDA compute capability at compile time. See tip below.
YOUR CUDA STORY AT SC12: Let us know how you use CUDA in 140 characters or less. Your submission will be considered for display on NVIDIA’s SC12 booth (#2217) and website.
Reply to the one-minute survey.


Using GPUs to Study the Human Brain
This week’s Spotlight is on Dr. Anders Eklund, a postdoc at the Virginia Tech Carilion Research Institute. Anders was a CUDA Spotlight in 2011 while a student at Linkoping University in Sweden. We caught up with him recently to learn about his current work. Read the interview here.
Dr. Anders Eklund


Robotic Bees?
The i09 website reports on a GPU computing project at the Universities of Sheffield and Sussex, where scientists "are hoping to create the first accurate computer simulation of a honey bee brain - and then upload it into an autonomous flying robot…. The researchers hope a robotic insect could supplement or replace the shrinking population of honey bees." See:

Chemistry and Life Science
Equip@meso will hold its first-ever scientific event - "Chemistry and Life Science: Numerical Simulation to HPC" - on Oct. 18 at the University of Strasbourg. Equip@meso is a France-based initiative to create a network of supercomputers for calculation-intensive tasks. See:

F# on the GPU
At the "F#unctional Londoners" Meetup on Oct. 18, attendees will hear a talk by Daniel Egloff on F# on the GPU with Alea.CUDA. Dr. Egloff is the founder of QuantAlea. See:

HPC Solutions Workshop
Dell is holding an event called "Enabling Discovery with Dell HPC Solutions" on Oct. 25 in Baltimore, Maryland. The workshop is targeted to researchers and developers. Speakers include:
   • Tamas Budavari, Johns Hopkins University
   • Anup Mahurkar, University of Maryland Baltimore
   • Jeffrey B. Klauda, University of Maryland College Park
   • Kevin Hildebrand, University of Maryland College Park
   • Jonathan Bentz, NVIDIA

CUDA Consulting
Training, programming, and project development services are available from CUDA consultants around the world. To be considered for inclusion on list, email: (with CUDA Consulting in subject line).

GPU Computing on Twitter
For daily updates about GPU computing and parallel programming, follow @gpucomputing on Twitter.


Title: CUDA Implementation of Parallel Algorithms for Animal Noseprint Identification
Author: Vincent Stanley Dayes, San Diego State University
Advisor: William Root, San Diego State University


Ventana Medical Systems, a member of the Roche Group, is seeking a Senior Imaging Scientist with expertise in image processing and computer vision to work with a collaborative team of engineers and scientists on initiatives pertaining to digital pathology. Familiarity with CUDA a plus. See:


back to the top
Stay up to date by reading the NVIDIA blog:
GPU-Accelerated Computing Surges in Russia, by Sumit Gupta
Synerscope: Data Analysis for the Rest of Us, by Brian Caulfield


back to the top
Find a GPU Meetup in your location, or start one up. Upcoming meetings include:
Silicon Valley, Oct. 8
Paris, Oct. 18 (special full day event)
Brisbane, Oct. 25
New York, Oct. 30

Note: If you would be interested joining a GPU Meetup in Minneapolis, Minn., see:


back to the top
Detect CUDA Compute Capability at Compile Time
When nvcc compiles a .cu file, it defines the __CUDA_ARCH__ preprocessor macro to a value representing the CUDA Compute Capability. Section 6.7.4 of the NVCC documentation ("CUDA Compiler Driver NVCC," included with the CUDA Toolkit) states that:

…the architecture identification macro __CUDA_ARCH__ is assigned a three-digit value string xy0 (ending in a literal 0) during each nvcc compilation stage 1 that compiles for compute_xy. This macro can be used in the implementation of GPU functions for determining the virtual architecture for which it is currently being compiled. The host code (the non-GPU code) must not depend on it.

As an example, this function uses the macro to only call Compute Capability 2.0 functions on devices that support them:
#if __CUDA_ARCH__ >= 200
    int b = __ballot();
    int p = popc(b & lanemask);
    // do something else for earlier architectures
(Source: Stack Overflow)


back to the top


- GPU Acceleration in Bioinformatics
Oct. 9, 2012, London, U.K.
Sponsored by Institute for Cancer Research and NVIDIA

- OpenACC Directives with PGI Accelerator Compilers (Webinar)
Oct. 11, 2012
Taught by Michael Wolfe, PGI
Note: Final webinar in a 3-part series

- OpenACC Workshop (RWTH Aachen University)
Oct. 11-12, 2012, Aachen, Germany
Taught by Timo Stich, NVIDIA and Sandra Wienke, RWTH Aachen University

- Nsight Eclipse: High Productivity IDE for CUDA on Linux, MacOS (Webinar)
Oct. 12, 2012, 10:00 am IST

- Portability, Scalability, Numerical Stability in Accelerated Kernels (Webinar)
Oct. 16, 2012, 9:00 am PT
Speaker: John Stratton, University of Illinois at Urbana-Champaign

- OpenACC Workshop (National Science Foundation, USA)
Oct. 16-17, 2012
Hosted by Pittsburgh Supercomputer Center, Natl. Inst. for Computational Sciences, Georgia Tech
Note: Telecast to 10 satellite sites around the USA

- Dell HPC Solutions Workshop
Oct. 25, 2012, Baltimore, Maryland
Note: Includes talks by researchers, complimentary lunch

- GPU Accelerated Applications and Academic Research
Oct. 31, 2012, 10:00 am PT
Speaker: Devang Sachdev, NVIDIA

- 4-Day CUDA Training Course (Acceleware)
Nov. 6-9, 2012, Houston, Texas

- SC12
Nov. 10-16, 2012, Salt Lake City, Utah

- GPUs in the Cloud
Dec. 3-6, 2012, Taipei, Taiwan

- Many-Core Developer Conference (UKMAC 2012)
Dec. 5, 2012, University of Bristol, UK


- GPU Tech Conference
March 18-21, 2013, San Jose, Calif.

(To list an event, email:



– CUDA 5:
– CUDA 5 survey:
– Nsight:
– CARMA (pre-register):

CUDA on the Web

– GPU Test Drive:
– Learn more about CUDA:
– Follow CUDA & GPU Computing on Twitter:
– Network with other developers:
– Stay tuned to GPGPU news and events:
– Newsletter archive:
– CUDA Spotlights:


CUDA® is a parallel computing platform and programming model invented by NVIDIA. It enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU). NVIDIA provides a complete toolkit for programming on the CUDA architecture, supporting standard computing languages such as C, C++ and Fortran. Send comments and suggestions on the newsletter to
You are receiving this email because you have previously expressed interest in NVIDIA products and technologies. Click here to opt in specifically to CUDA: Week in Review. NVIDIA - World Leader in Visual Computing Technologies
Feel free to forward this email to customers, partners and colleagues.

Copyright © 2012 NVIDIA Corporation. All rights reserved. 2701 San Tomas Expressway, Santa Clara, CA 95050.