CUDA GPGPU Parallel Computing Newsletter - Issue 57

GPU-Accelerated Science and Computing
This week’s spotlight is on Dr. Vincent Natoli, president and founder of Stone Ridge Technology. Dr. Natoli is a computational physicist with 20 years of experience in high performance computing. Previous roles include senior physicist at ExxonMobil Corporation and Technical Director at High Performance Technologies Inc. (HPTi). Here’s a preview of our interview:

NVIDIA: Vincent, tell us a bit about Stone Ridge Technology.

Vincent: Stone Ridge provides products and services to the HPC market. I had the idea for the company back in 2002 and started it full time in 2005. I wanted to build a company that solves difficult problems in science and engineering on leading edge hardware platforms. It’s the border between science and computing which I love.

NVIDIA: What services do you provide?

Vincent: Most of our customers are in the oil and gas industry. We port, optimize and develop from scratch high performance technical codes for some of the biggest corporations in the world.

NVIDIA: What advice do you have for developers considering GPU computing?

Vincent: I would say that if performance is important to you then GPU computing is well worth a try. Take a look at CUDA Zone to find a code similar to your own to see what kind of performance you can expect from the GPU. Profile your code and try porting the most significant hotspot first. The potential gain is well worth a few weeks of investigation. Along the way you will learn a lot about your code.

- Read the complete interview here

(Would you like to be in the CUDA Spotlight? Email cuda_week_in_review@nvidia.com)

Microsoft and NVIDIA Host C++ GPU Computing Event

back to the top

Silicon Valley engineers met at NVIDIA’s headquarters this week to learn about C++ technologies in heterogeneous environments (CPU + GPU). The evening kicked off with a talk by luminary Herb Sutter, chief architect for Windows C++ at Microsoft, speaking about plans to take advantage of GPU acceleration with C++ AMP. NVIDIA experts (Justin Luitjens, Jared Hoberock, Stephen Jones) provided an overview of GPU computing and CUDA to developers from Cisco, eBay, Facebook, Google, MTV, Visa and other companies.

Quantum Chromodynamics and CUDA
A recent story in "International Science Grid This Week" points out that researchers worldwide have access to a growing library of quantum chromodynamics (QCD) software ported to run on GPUs. "Lattice QCD is in many ways ideal for GPUs, since the bulk of QCD computations can be broken down into many parallel independent calculations...," said Mike Clark, a physicist at Harvard.

Microway Wins Best New Technology Award
Microway, a Tesla Preferred Partner, won the Best New Technology award at the Life Sciences Technology Insight Summit in Boston. Microway’s Tesla-based solutions are delivered fully integrated with CUDA and Linux or Windows. At the Summit, two Microway customers presented their work:
- Dr. Preston Moore of the University of the Sciences in Philadelphia discussed GPUs and the
physicochemical properties of biological membranes.
- Dr. John Wolfgang of Mass General Hospital spoke about GPU-based radiotherapy visualization
for cancer patient treatment.

Russia Accelerates Scientific Innovation with GPU Supercomputers
Moscow State University is upgrading its Lomonosov system with Tesla GPUs to be one of the world's fastest supercomputers. The upgraded system couples 1,554 NVIDIA Tesla X2070 GPUs with the same number of quad-core CPUs, to deliver an expected 1.3 petaflops of peak performance.

CST Adds New HPC Capabilities
Computer Simulation Technology (CST) announced CST MICROWAVE STUDIO v2011 SP3, with support for Tesla GPUs. CST MWS enables fast analysis of high frequency devices such as antennas, allowing shorter development cycles and virtual prototyping.

MATLAB Updates
MATLAB made the prerelease of MATLAB R2011b available to existing MATLAB Parallel Computing Toolbox licensees, adding a host of new GPU features. A new MATLAB-related post on the Accelereyes blog discusses "Chan-Vese Active Contours on the GPU." Active Contours are mathematical models that enable detection of objects within images.

ISC 2011 Content Available for Download
Visit the NVIDIA ISC event page to download presentations from the show including the industry briefing on GPU computing and the CUDA tutorial.

Publications from NVIDIA
Check out the new CUDA 4.0 Library Performance Overview and white paper on Sparse Triangular Linear Systems.

Supercomputing ‘Energy Wall’ –NVIDIA Helps Scale It – By Sumit Gupta
ISC 2011: Europe Flexes Its GPU Supercomputing Muscles – By Andy Walsh
CUDA Now Available For Multiple x86 Processors – By Sumit Gupta
Microsoft Going All-in On GPU Computing – By Sanford Russell

back to the top

NEW: Each week we highlight sessions from GTC 2010 and SC10. Here are our picks for this week:

back to the top

      State of the Art in GPU Data-Parallel Algorithm Primitives (GTC 10)
      Mark Harris - NVIDIA
      http://nvidia.fullviewmedia.com/gtc2010/0921-a5-2084.html (streaming)

      Large-Scale GPU Computing of Multi-Phase Flow (SC10)
      Wei Ge - Institute of Process Engineering, Chinese Academy of Sciences
      www.nvidia.com/content/PDF/sc_2010/theater/Ge_SC10.pdf

NEW: Varian Medical Systems in Palo Alto, Calif., is seeking a graduate student intern with CUDA programming experience. Familiarity with CBCT (cone beam computed tomography) reconstruction algorithms is required.
- See: www.varian.com/us/corporate/careers/

back to the top

If you are travelling to any of these cities, feel free to drop in. Visitors are welcome.
Boston Meetup – July 7, 6:00 pm
Austin Meetup – July 18, 6:00 pm
Silicon Valley Meetup – July 18, 6:15 pm
Australian Tele-Meetup – July 21, 4:00 pm
New York Meetup – July 28, 6:00 pm

back to the top

Join us on July 12 for a special webinar on CUDA Global Memory Usage and Strategy, followed by a live Q&A session. In the webinar, Justin Luitjens of NVIDIA will describe the CUDA Memory Model, focusing on global memory and strategies to leverage available bandwidth. Other upcoming webinars include CUDA Shared Memory & Cache (July 19) and CUDA Warps and Occupancy Considerations (July 26).
- See schedule here: www.nvidia.com/webinars

back to the top

July 2011	back to the top

- NEW: Parallel Nsight and Microsoft Visual Studio Webinar

July 5, 2011 (in Russian)

http://blogs.msdn.com/b/rudevnews/archive/2011/06/23/gpu-visual-studio.aspx

- CUDA GPU Programming Workshop

July 2-3, 2011 Berlin, Germany

http://cuda.eventbrite.com/

- 2011 World Congress on Engineering (WCE 2011)

July 6-8, 2011, London, England

http://www.iaeng.org/WCE2011/ICPDC2011.html

- Lattice 2011 Conference

July 11-16, 2011, Lake Tahoe, California

https://latt11.llnl.gov/

- NEW: CUDA Global Memory Usage and Strategy Webinar

July 12, 2011, 10:00 am pacific

https://www2.gotomeeting.com/register/768504595

- Genetic and Evolutionary Computation Conference (GECCO)

July 12-16, 2011 Dublin, Ireland

www.sigevo.org/gecco-2011/index.html

- World Congress in C.S., Computer Engineering, Applied Computing (WORLDCOMP’11)

July 18-21, 2011, Las Vegas, Nevada

www.world-academy-of-science.org/worldcomp11/ws

- Application Accelerators in High Performance Computing (SAAHPC 2011)

July 19-21, 2011, Univ. of Tennessee, Knoxville, Tennessee

http://illinois.edu/lb/article/2100/45663

- GTC Workshop Japan

July 22, 2011, Tokyo

Hosted by NVIDIA, with the Tokyo Institute of Technology

http://www.nv-jp-event.jp/gtc-workshop2011/

- NEW: Banking with GPGPUs: Increased Performance, Lowered Costs (Training Workshop)

July 25-27, 2011, London, UK

Note: Taught by experts from Excelian and Cranfield University

Contact: james.heath (at) excelian.com

Aug. – Sept. 2011

- LAMMPS Users' Workshop

Aug. 9-1, 2011, Albuquerque, New Mexico

Learn more about LAMMPS: http://lammps.sandia.gov/index.html

- NEW: Banking with GPGPUs: Increased Performance, Lowered Costs (Training Workshop)

August 8-11, 2011, London, UK

Note: Taught by experts from Excelian and Cranfield University

Contact: james.heath (at) excelian.com

- Advanced Numerical Methods on GPUs

Mini-symposium at ENUMATH 2011

Sept. 5-9, 2011, Univ. of Leicester, Leicester, UK

http://www2.le.ac.uk/departments/mathematics/research/enumath2011/minisymposia

- Parallel Processing and Applied Mathematics (PPAM 2011)

Sept. 11-14, 2011, Torun, Poland

Note: Scientific Computing with GPUs tutorial, incl. session by Tim Schroeder, NVIDIA

http://ppam.pl/tutorials/21

- Geospatial Summit

Sept. 13-14, 2011, Herndon, Virginia

http://cfp.foseinstitute.org/gis2011

- SPIE Conference on High-Performance Computing in Remote Sensing

Sept. 19-22, 2011, Prague, Czech Republic

http://spie.org/rs11

- SEG (Society of Exploration Geophysicists) Annual Meeting

Sept 18-23, 2011, San Antonio, Tex.

http://www.seg.org/events/annual-meeting/sanantonio2011/am2011techprogram

(To list an event, email: cuda_week_in_review@nvidia.com)

Downloads	back to the top
– CUDA 4.0: www.nvidia.com/getcuda – Parallel Nsight: http://developer.nvidia.com/nvidia-parallel-nsight
Webinars
– CUDA: http://developer.nvidia.com/gpu-computing-webinars – Parallel Nsight: http://developer.nvidia.com/developer-webinars

CUDA Registered Developer Program

– Sign up: www.nvidia.com/paralleldeveloper

CUDA GPUs

– List of CUDA-enabled GPUs: www.nvidia.com/object/cuda_gpus.html

CUDA on the Web

– See previous issues of CUDA: Week in Review: http://is.gd/cBXbg
– Follow CUDA & GPU Computing on Twitter: www.twitter.com/gpucomputing
– Network with other developers: www.gpucomputing.net
– Stayed tuned to GPGPU news and events: www.gpgpu.org
– Learn more about CUDA on CUDA Zone: www.nvidia.com/cuda
– Check out the NVIDIA Research page: www.nvidia.com/research

CUDA Recommended Reading

– Future of Computing Performance: http://bit.ly/hYqH2H
– Supercomputing for the Masses, Part 21: http://is.gd/Fj56gf
– CUDA books: http://www.nvidia.com/object/cuda_books.html

CUDA Recommended Viewing

– The Third Pillar of Science: www.nvidia.com/object/race-for-better-science.html

– GTC 2010 presentations: www.nvidia.com/gtc

– SC10 presentations: www.nvidia.com/object/sc10_theater.html

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: cuda_week_in_review@nvidia.com

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.