CUDA: Week in Review
Friday, May 7, 2010, Issue #20 - Newsletter Home  
Welcome to this week’s issue of "CUDA: Week in Review," a weekly newsletter for the worldwide CUDA and GPU Computing community. Contact us at: Follow us on Twitter:

Reminder: The GPU Technology Conference (GTC), Sept. 20-23, is accepting proposals for GPU-related sessions on topics ranging from astronomy to energy exploration to video processing. The deadline is June 1. Learn more: and
Tesla M2050 Arrives
This was an important week for developers looking to scale applications across multiple GPUs, as the first server products based on Tesla M2050 GPU Computing modules were announced. Tesla M2050 systems are being offered by Appro, Supermicro and Tyan. In addition, Bright Computing is providing GPU cluster management software with new capabilities.
Watch Andy Keane, NVIDIA general manager, explain why Tesla is so exciting:
CUDA, Finance and the City of Lights
The Global Derivatives Trading & Risk Management show will be held May 17-21 in Paris. The growing role of GPUs in finance will be highlighted in sessions such as:
– Simon Rees, Barclays Capital: "Large Scale Monte Carlo Loss Simulation using GPUs"
– Prof. Claudio Albanese, King’s College, London: "High Performance Pricing"
– Dr. Curtis Randall, SciComp: "Automatic GPU Computing for Derivative Pricing Models"
NVIDIA partners Murex, Numerical Algorithms Group (NAG), and SciComp will exhibit. To learn more, see:
Speeding Up GIS (Geographic Information Systems)
GIS experts around the world have been monitoring the huge oil spill in the Gulf of Mexico. GIS technology helps specialists interpret oil slicks in order to assist in response, planning and damage assessment. Incogna GIS of Ontario, Canada has developed an application called "GIS Image Analysis On-Demand," which uses GPU-based image analysis techniques for computationally-intensive tasks such as surface classification. These techniques leverage CUDA to process three days worth of data in less than one hour. The core technology behind Incogna GIS is ICRE (Image Content Recognition Engine), a cloud-computing, GPU-based computer vision system. See:
Update on GPU-Accelerated Data Mining
A few weeks ago we told you about Jedox’s Online Analytical Processing (OLAP) tools called Palo Suite. The Jedox offering is explored in more detail in a blog post by tech writer Steve Wildstrom, who says: "Data mining may not seem to be a natural fit for parallel processing. Yet at least one data mining software maker is scoring impressive performance gains using GPU processing for OLAP, a technique for taking a deep dive into a subset of what may be a very large database." See blog post here:
Every week we learn about new CUDA and GPU Computing courses being taught worldwide. Here are a few new ones:
Course name: Graphic Processors in Computational Applications
Level: Undergrad and graduate
Location: Warsaw University of Technology, Poland
Instructor: Prof. Krzysztof Kaczmarski
Course name: CUDA and Scientific GPU Computing
Level: PhD
Location: Technical University of Denmark
Instructor: Prof. Allan Engsig-Karup
Course name: Programming and Tuning Massively Parallel Systems - Summer School
Level: Beginner through advanced
Location: Barcelona Supercomputing Center, Universitat Politecnica de Catalunya, Spain
Instructors: Dr. Wen-mei Hwu, Dr. David Kirk
Note: One-week course. Applications due May 20.
New on CUDA Zone: Fast Human Detection with Cascaded Ensembles
(Master's thesis submission, Dept. of Electrical Engineering and Computer Science, MIT)
Extract: "This thesis addresses the problem of object detection from images, in particular the detection of people. As digital cameras become more widespread, the volume of available data to digital camera owners reach such a point that digital content management presents itself as a problem. In our work, we use the NVIDIA CUDA framework. CUDA is the computing platform that enables developers to code parallel algorithms through industry standard languages. The CUDA programming model acts as a platform for massively parallel high performance computing by providing a direct, general-purpose C language interface (‘C for CUDA’) to the programmable multiprocessors on the GPUs. When implemented on this platform, we observed a significant speed up in our cascade detector‘s performance." Authored by Berkin Bilgic, MIT student. See:
CUDA Zone Submissions
Have a CUDA-related paper, research, or app? Show it on CUDA Zone:
HCL Technologies in India is looking for two CUDA developers for their Bangalore office. "Must-have" requirements include CUDA development experience and C/C++ expertise. Experience with OpenMP/MPI, OpenGL/DirectX, C#/.net, and compiler technologies a plus. HCL is a leading global IT services company, working with clients in areas that impact and redefine the core of their businesses. HCL leverages its extensive global offshore infrastructure and network of offices in 26 countries for industry verticals including Financial Services, Manufacturing, Consumer Services, Public Services and Healthcare.
– To contact HCL directly about these positions, email:
– For more info on HCL, see:
GPU Computing Webinars (CUDA C, OpenCL, Parallel Nsight and more…)
See upcoming May webinars:
CUDA Training
– SagivTech CUDA Training, May 10-12, Ra’anana, Israel:
– Acceleware-Certified CUDA Training, May 19-20, Silicon Valley:
CUDA and Academia
Over 340 universities are teaching CUDA and GPU Computing courses.
– See the list:
– GPU Computing in the Oil & Gas Industry (Microsoft/NVIDIA)
May 12, Houston

– NEW: Global Derivatives Trading & Risk Management
May 17-21, Paris

– ISC ´10 GPU Computing Workshops
May 30, Hamburg, Germany

– Parallel Execution of Sequential Programs on Multi-Core Architectures
June 20, France

– NEW: GPGPU Briefing for Financial Services (Microsoft/NVIDIA)
June 21, New York City

– GPUs in Chemistry and Materials Science
June 28-30, Univ. of Pittsburgh

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

– Symposium on Chemical Computations on GPGPUs
Aug. 22-26, Boston

– Unconventional High Performance Computing 2010 (UCHPC 2010)
Aug. 31-Sept. 1, Italy

– GPU Technology Conference 2010
Sept. 20-23, San Jose, Calif. (now accepting proposals from industry and academia)

(To list an event, email:

GPU-accelerated linear algebra library from EM Photonics:
NVIDIA Parallel Nsight
Download the Parallel Nsight Beta:
CUDA Toolkit
Download CUDA Toolkit 3.0:
CUDA Documentation
Download developer guides and documentation:
CUDA Books
– Programming Massively Parallel Processors by D. Kirk, W. Hwu:
– See additional books here:
– Follow CUDA & GPU Computing on Twitter:
– Network with other developers:
– Stayed tuned to GPGPU news and events:
– Learn more about CUDA on CUDA Zone:
– 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.

See previous issues of CUDA: Week in Review:

Send comments and suggestions 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.

Feel free to forward this email to customers, partners and colleagues.

Copyright © 2010 NVIDIA Corporation. All rights reserved. 2701 San Tomas Expressway, Santa Clara, CA 95050.