CUDA: WEEK IN REVIEW

Welcome to CUDA: WEEK IN REVIEW, a news summary for the worldwide CUDA, GPGPU and parallel programming community.

Register for GTC 2013, the world’s premier CUDA developer conference (Discount code for newsletter readers: GM10CD)

CUDA Newsletter Reader Survey: We want to hear from you Please take a minute to fill out our reader survey so we can improve the newsletter in 2013. Thanks!

CUDA Tech Tip: Flush denormals with confidence It pays to understand numerical requirements of your code. You may be able to get a big performance boost by using the compiler command line option “–ftz=true” if you use operations like sqrtf() and rsqrtf() and you don’st need denormal support. Learn more on Parallel Forall.

CUDA SPOTLIGHT
GPU-Accelerated Brownout Simulations This week’s Spotlight is on Monica Syal, an aerospace engineer at Sunnyvale, Calif.-based Advanced Rotorcraft Technology (ART). Monica's responsibilities include development of a real-time pilot-in-the-loop rotorcraft brownout simulation and mitigation methodology for flight simulator applications. Read our interview with Monica Syal
CUDA NEWS
CUDA 5 Performance Report Now Available Learn how your application can run faster on a Tesla K20 GPU Accelerator, simply by dropping in a GPU-accelerated library.    • Download the new CUDA 5 Performance Report (PDF 1.06 MB)    • Register for CUDA 5 Math Library Performance Overview webinar, Wed., Feb. 6 New: SIMD-in-a-Word Source Code NVIDIA posted new source code for CUDA Registered Developers which provides “SIMD-within-a-word” functionality, useful for simulation and image processing codes.    • See the DevTalk Forum announcement    • Join the CUDA Registered Developers program
CUDA EDUCATION	back to the top
Udacity Parallel Programming Course Launches This Week! Sign up for Intro to Parallel Programming. Learn more about Udacity. AccelerEyes Training Courses Announced AccelerEyes has announced CUDA trainings in the U.S. for 2013. Visit AccelerEyes Training.
CUDA JOB OF THE WEEK	back to the top
Crane, the well-known banknote and stationery company, seeks a Software and Design Engineer. Experience with AutoCAD, Python, C, C++ and CUDA C preferred.
FROM THE BLOGOSPHERE	back to the top
New on the NVIDIA blog: Just Mix Water with GPUs..., by Roy Kim New on the Parallel Forall blog: Using Shared Memory in CUDA C/C++, by Mark Harris (Subscribe to the Parallel Forall RSS feed)
GPU MEETUPS	back to the top
Find a GPU Meetup in your location, or start one up. Upcoming meetings include: Paris, Feb. 18 Silicon Valley, Feb. 25 New York, Feb. 27 Boston, Feb. 28 Brisbane, Feb. 28
CUDA CALENDAR	back to the top
February HPC Advisory Council Stanford Conference   Feb. 7-8, 2013, Stanford, Calif. Open to the public Note: NVIDIA's Sarah Tariq to present on scaling applications on large GPU clusters Intro to CUDA + OpenACC Course   Feb. 12-14, 2013, Ukraine, Lviv Sponsored by Applied Parallel Computing LLC Contact: akomissarov@nvidia.com Winter GPU Computing School (Moscow State University)   Feb. 20-22, 2013, Moscow, Russia 2-Day CUDA Training (AccelerEyes)   Feb. 25-26, 2013, Houston, Texas March Intro to CUDA + OpenACC Course   March 4-5, 2013, Hamburg, Germany Sponsored by Applied Parallel Computing LLC Contact: akomissarov@nvidia.com 2-Day CUDA Training (AccelerEyes)   March 4-5, 2013, Washington, DC 4-Day CUDA Course - Oil & Gas (Acceleware)   March 12-15, 2013, Houston, Texas Instructor: Dr. Kelly Goss HPC Advisory Council   March 13-15, 2013, Lugano, Switzerland GPU Tech Conference (GTC 2013)   March 18-21, 2013, San Jose, Calif. Special code for CUDA newsletter readers: GM10CD 2-Day CUDA Training (AccelerEyes)   March 25-26, 2013, Los Angeles, Calif. April-May 4-Day CUDA Course (Acceleware)   May 7-10, 2013, San Francisco, Calif. Instructor: Dr. Kelly Goss (To list an event, email: cuda_week_in_review@nvidia.com)
CUDA RESOURCES	back to the top
GPU-Accelerated Apps List of 200+ popular GPU-accelerated scientific and research applications (PDF 402KB). GPU Test Drive Want to try Tesla K20 for free? Sign up here. CUDA Documentation The new CUDA documentation site includes release notes, programming guides, manuals and code samples. CUDA Education Udacity Course Coursera Course NVIDIA Developer Forums Join us on the NVIDIA DevTalk forums to share your experience and learn from other developers. You can also ask questions on Stack Overflow, using the ‘cuda’ tag. CUDA Consulting Training, programming and project development services are available from CUDA consultantsaround the world. To be considered for inclusion on list, email:cuda_week_in_review@nvidia.com (with CUDA Consulting in subject line). GPU Computing on Twitter For daily updates about GPU computing and parallel programming, follow @gpucomputing on Twitter. Downloads CUDA 5 CUDA 5 survey Nsight CUDA on the Web CUDA Spotlights CUDA Newsletters CUDA Zone GPU Test Drive GPUComputing.net GPGPU.org