CUDA Week in Review Newsletter Homepage
Mon., Aug. 27, 2012, Issue #78 Newsletter Home
Welcome to CUDA: WEEK IN REVIEW, a news summary for the worldwide CUDA, GPGPU and parallel programming community.


Want to time the performance-critical regions in your application? NVIDIA Nsight can do it very efficiently, without the need to write new code. Simply label the regions using the NVIDIA Tools Extensibility Library (NVTX), a simple cross-platform C API. When activated, Nsight logs NVTX calls with minimum overhead. NVTX is included in the Visual Studio and Eclipse editions of Nsight, with documentation in the Profiler User’s Guide in the CUDA Toolkit doc folder.


GPU-Accelerated Radiation Physics
This week's Spotlight is on Dr. Michael Bussmann of HZDR, a large-scale research facility based in Dresden, Germany.

Michael leads a "Junior Research Group" in the field of Computational Radiation Physics. In our interview, you'll learn how GPU computing dramatically speeds up his team's simulations.
Dr. Michael Bussmann


Download CUDA 5 RC
The CUDA 5 Release Candidate (RC) is now publicly available, with advanced features such as GPU Object Linking.
      •  Download CUDA 5 RC:
      •  Register for CUDA 5 webinar:

CUDA Consulting
Did you know that training, programming, and project development services are available from CUDA consultants around the world?
      •  See partial list here:
      •  To be considered for inclusion on list, email: (with CUDA
          Consulting in subject line)

New CUDA Research Centers
NVIDIA has named three new CRCs with the following areas of focus:
      •  University of Vigo, Spain: Multi-GPU strategies for the open source DualSPHysics model
      •  University of Southern California ISI: High performance computing in the cloud
      •  Wake Forest University: GPU-optimized molecular dynamics simulations

GPUs and Databases
A workshop on GPUs and databases will be held at ADBIS on Sept. 17 in Poznan, Poland. Talks include:
      •  Applying CUDA Technology in DCT-based Method of Query Selectivity Estimation
               ο  By Dariusz Rafat Augustyn and Sebastian Zederowski
      •  Processing of Range Query Using SIMD and GPU
               ο  By Pavel Bednar, Petr Gajdos, Michal Kratky and Peter Chovanec
      •  Towards Optimization of Hybrid CPU/GPU Query Plans in Database Systems
               ο  By Sebastian Bress, Eike Schallehn and Ingolf Geist
      •  Thrust and CUDA in Data Intensive Algorithms
               ο  By Krzysztof Kaczmarski and Pawel Rzazewski

GPUs in the Cloud
NVIDIA's partners in Europe, Asia, India and North America are offering a free "GPUs in the cloud" test drive for GPU accelerated applications. Molecular dynamics apps such as AMBER, NAMD and LAMMPS are pre-loaded so there is no need to set up any hardware or software - simply run your models and measure the speedup. You can also load other apps or your own code. Sign up today at

Update from AccelerEyes
AccelerEyes, developer of ArrayFire and Jacket, has posted new case studies related to Parallelized Gene Predictors and Options Pricing.
      •  See AccelerEyes case studies at
      •  Sign up for AccelerEyes webinars at

New Release from Xcelerit
Xcelerit, provider of cross-platform acceleration tools for financial services, engineering, and research, announced Xcelerit SDK 2.0. The Xcelerit SDK is designed to boost performance of compute-intensive applications while preserving programmer productivity. Learn more:


Congrats to Siu Kwan Lam of San Jose State University on his M.S thesis titled "Scaling CUDA for Distributed Heterogeneous Processors." Advisor: Dr. Donald Hung, Dept. of Computer Engineering. Date: May 2012. Read it here:


HeartFlow, a cardiovascular diagnostics company in Redwood City, Calif., is seeking a software developer for advanced image processing of medical data. The candidate will be responsible for development of user facing apps on Windows and automated algorithms in Linux. See:


back to the top
Stay up to date by reading the NVIDIA blog:
Unleash Legacy MPI Codes with Kepler's Hyper-Q, by Peter Messmer
CUDA Centers Top 200, by Chandra Cheij
NVIDIA Hits Emission Standard, by Tonie Hansen


back to the top
Find a GPU Meetup in your location, or start one up. Upcoming meetings include:
Sydney, Australia GPU Meetup, Aug. 28
Paris GPU Meetup, Sept. 5
Silicon Valley GPU Meetup, Sept. 17
Boston GPU Meetup, Sept. 26

(Want to start a GPU Meetup? Contact


back to the top


- Summer School of Advanced Computing (CASPUR)
Aug. 27-Sept. 7, 2012, Rome, Italy

- Latin America Conference on HPC (CLCAR)
Aug. 27-31, 2012, Panama

- Nsight Eclipse Edition: High Productivity IDE (Webinar)
Aug. 28, 2012, 10:00 a.m. PT

- Introducing CUDA 5 (Webinar)
Aug. 29, 2012, 9:00 a.m. PT

- Parallel Computing Summer School (CINECA)
Sept. 3-14, 2012 (in English), Bologna, Italy

- GPU Object Linking: Usage and Benefits (Webinar)
Sept. 5, 2012, 10:00 a.m. PT

- Computational Genomics… Using OpenACC (Webinar)
Sept. 6, 2012, 9:00 a.m. PT

- Workshop on GPUs and Databases (ADBIS)
Sept. 17, 2012, Poznan, Poland

- MATLAB for CUDA Programmers (Webinar)
Sept. 19, 2012
Presenters: Daniel Armyr, MathWorks and Jonathan Bentz, NVIDIA

- Running Risk on GPUs (Webinar)
Sept. 19, 2012, 9:00 am PT
Presenter: Tim Wood, Quantitative Analyst, ING Bank nv

- Genomics MATLAB Apps… with Jacket (Webinar)
Presented by AccelerEyes
Sept. 19, 2012, noon PT

- PGI Accelerator with OpenACC Directives
Sept. 26, 2012, Houston, Texas


- 4-Day CUDA Training Course (Acceleware)
Oct. 2-5, 2012, San Jose, Calif.

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


- GPU Tech Conference
March 18-21, 2012, San Jose, Calif.

(To list an event, email:



– Nsight:
– CARMA (pre-register):

CUDA on the Web

– Learn more:
– 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

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