CUDA: Week in Review
Friday, July 2, 2010, Issue #28 - 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:
GPUs Boost Supercomputers’ Energy Efficiency
The Green500 List of the world’s most energy-efficient supercomputers was published this week. Agam Shah of PCWorld writes: "Supercomputers that mix CPUs with graphics processors made their mark on the Green500 list….Two of the top eight green supercomputers are new entrants from China, and combine graphics processors from NVIDIA with Intel’s CPUs." Read Agam’s article here:
Hot off the Press: GPU Technology Conference - Sessions Preview
The GTC team has identified a preliminary list of sessions to be presented in September. While we are still reviewing all the submissions, here are a few confirmed presentations to whet your appetite:

– Bridging GPU Computing and Neuroscience to Build Large-Scale Face Recognition on
       Nicolas Pinto, PhD Student, MIT

– Using GPUs to Follow the Sun’s Latest Moves
       Mark Cheung, Sr. Physicist, Lockheed Martin Solar & Astrophysics Lab

– Shockingly Fast and Accurate CFD Simulations
       Timothy Warburton, Associate Professor, Rice University

– Banking on Monte Carlo… and Beyond
       Dr. Ian Reid, Numerical Algorithms Group

– WebGL: Bringing 3D to the Web
       Vladimir Vukicevic, Principal Engineer, Mozilla Corporation
More GTC Info
GTC 2010 sponsors include Adobe, HP, Microsoft, Supermicro, PNY, Dell, Cooley, Synnex, Next IO, GE Intelligent Platforms, AMAX, SGI, Appro and Sutter Hill Ventures, and media partner Dr. Dobbs.
– Register for GTC today as space is limited (code: GMCUDANEWS10)
– For list of sessions, see the GTC agenda page
– Join the GTC Facebook fan page and GTC email update list
GPUs and Science
Researchers gathered this week at the University of Pittsburgh’s Center for Simulation and Modeling for a symposium on GPUs in chemistry and materials science. Here they discussed state-of-the-art GPU-enabled applications such as AMBER, VMD, Q-CHEM and Terachem. Dr. Ross Walker of the San Diego Supercomputer Center reviewed the native capabilities of AMBER 11 utilizing GPUs. Dr. Scott Le Grand of NVIDIA gave a presentation on AMBER + MPI performance developments. A related event, the American Chemical Society meeting, will take place on Aug. 22-26 in Boston. For information on NVIDIA and biochemical research, see:
New on CUDA Zone: Software Transactional Memory for GPUs
Authors: Daniel Cederman, Philippas Tsigas, Muhammad Tayyab Chaudhry; Chalmers University of Technology, Sweden
Application Domain: Programming Tools
Extract: "Software Transactional Memory (STM) allows you to write sequential code that is executed concurrently without using locks. This greatly simplifies the implementation of parallel programs. The simplicity, however, comes at the price of performance. With this work we try to shed some light on the performance price of STMs on CUDA…. The behavior of STMs when the contention to shared data is low might be acceptable. But when the contention is high, it is better to try to design the program to be data parallel, which is where CUDA excels, and/or to use lock-free data structures." See:
CUDA Zone: Have a CUDA-related app or paper? Let us know when you post it on CUDA Zone and we’ll send you a CUDA t-shirt!
Distinct, a consultancy and outsourcing provider, uses advanced data modeling techniques to improve operational performance and create growth. Clients include banking, insurance and healthcare companies in Ireland and the U.K. The prospective role will involve development of Distinct’s analytical platform using GPU-based architectures. Requirements include experience in GPU programming, architectures and in particular, CUDA. Contact
Microsoft Launches DirectCompute Lecture Series
Microsoft launched a series of lectures focused on DirectCompute. A component of DirectX, DirectCompute is Microsoft’s GPGPU programming solution for Windows. Videos of the lectures are posted online.

– Robert Hess of Microsoft hosts a roundtable with thought leaders, including Eric Young
   of NVIDIA

– James Fung of NVIDIA speaks on "DirectCompute GPU Optimizations and Performance"

– More info:
GPU Computing Webinars (CUDA C, OpenCL, Parallel Nsight and more…)
NVIDIA’s webinars are increasingly popular in the GPU computing community and represent a great way to get up to speed quickly. Here are several coming up in July:

– Intro to OpenCL Application Development with gDEBugger
   Tuesday, July 6, 9:00 a.m. pacific

– Rapid Prototyping and Visualization with OpenCL Studio
   Wednesday, July 7, 9:00 a.m. pacific

– Monitoring and Mapping GPU Clusters with Bright Cluster Manager
   (Presented by CEO and Founder of Bright Computing, Dr. Matthijs van Leeuwen)
   Thursday, July 8, 9:00 a.m. pacific
CUDA and Academia
– Over 350 universities are teaching CUDA and GPU Computing courses.
– The CUDA Center of Excellence (CCOE) Program recognizes universities expanding the
   frontier of parallel computing.
SagivTech Announces CUDA Course on Image Processing, Following GTC
SagivTech’s new CUDA course will take place Sept. 27-29 in the San Francisco area, following the GPU Technology Conference. This course is for CUDA developers looking for optimization for image processing on NVIDIA GPUs. Course instructors are Micha Feigin Almon, an experienced developer, mathematician and image processing specialist and Dr. Chen Sagiv, an applied mathematician and expert in image processing algorithms. This is a convenient option for GTC attendees. See

– Other CUDA training courses from SagivTech:
CUDA course: July 12-14, Ra’anana, Israel
GPU/Image Processing course: Aug. 2-4, Ra’anana, Israel
CUDA training from Acceleware
July 26-30, Cambridge, Mass: (with Microsoft)
Aug. 2-6, New York City: (with Microsoft)
Sept. 13-17, Calgary:
CUDA training from EMPhotonics
On-site standard and customized training programs
CUDA Research and Certification
For info on NVIDIA programs for GPU Computing developers, see:
July 2010

Programming and Tuning Massively Parallel Systems
July 5-9, Univ. Politecnica de Catalunya, Spain

NVIDIA Bio Workbench and AMBER 11
July 8, Imperial College London

Hybrid Cloud Computing with NVIDIA GPUs and ProActive Parallel Suite
July 8, INRIA, University of Nice, Sophia Antipolis, France

Symposium on Application Accelerators in High-Performance Computing
   (SAAHPC 2010)

July 13-15, University of Tennessee at Knoxville

Parallel Symbolic Computation 2010 (PASCO)
July 21-23, Grenoble, France

July 25-29, Los Angeles

CUDA Programming on GPUs
July 26-30, University of Oxford, U.K.

August 2010

Proven Algorithmic Techniques for Many-Core Processors
Aug. 2-6, Choice of onsite locations, or online

Virtual School of Comp. Science & Engineering
Aug. 2-6, choice of onsite locations

GPU Programming for Molecular Modeling
Aug. 6-8, Beckman Inst. for Advanced Science & Tech, UIUC

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)


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 18:
– Supercomputing for the Masses, Part 17:
– Supercomputing for the Masses, Part 16:
– Supercomputing for the Masses, Part 15:
CUDA Books
– Programming Massively Parallel Processors by D. Kirk, W. Hwu:
– See additional books here:
CUDA Toolkit
– Download CUDA 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.
CUDA Documentation
Download developer guides and documentation:
NVIDIA Parallel Nsight
– Download the Beta:
– Download the Release Notes:
– Check out the NVIDIA Research site:
– 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:
– Watch CUDA on YouTube:
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.