CUDA: Week in Review
Friday, March 12, 2010, Issue #12 - Newsletter Home  
Welcome to this week’s issue of "CUDA: Week in Review." Questions or suggestions? Email us at
Bringing GPU Power to Engineering Computations
MATLAB from MathWorks is a high-level language and interactive environment that enables developers to perform computationally intensive tasks faster than with traditional programming languages. Now there’s a new tool called "AccelerEyes Jacket" that provides a way for MATLAB users to easily tap into the power of CUDA GPUs for advanced computations. Tech editor Peter Varhol of Desktop Engineering writes that the AccelerEyes product "acts as kind of a traffic cop for executing code‚ diverting code to run on the GPU when appropriate." Read the full story:
CUDA for Optical Character Recognition
Russia-based Cognitive Technologies is an OCR (optical character recognition) software developer. Its Cognitive Passport product is used for the scanning and recognition of documents such as passports and ID cards in environments where speed and accuracy are critical‚ such as busy airports where thousands of documents are processed daily. Recently‚ Cognitive released new software that leverages CUDA, achieving a 30X speedup in the recognition phase of the application on a GPU vs. a CPU. Learn more here:
GPUs‚ Open Source‚ and Finance
Professor Mark Joshi is a well-known author and quant (note: a "quant" designs and implements mathematical models for the pricing of derivatives‚ assessment of risk‚ or predicting market movements). Mark recently launched a new open source project called Kooderive. His objective is to utilize CUDA-enabled GPUs to produce fast Monte Carlo pricing models for financial derivatives. He has already developed a fast Monte Carlo path generator for pricing exotic interest rate derivatives. The next stage will be to integrate with existing code in the QuantLib open source project. On June 2‚ Mark will teach a 3-day course on pricing exotic interest rate derivatives at the Institute of Physics in London. For more info‚ see:
New on CUDA Zone: Random Linear Network Coding on GPUs
Extract: "Random linear network coding has recently been widely applied in peer-to-peer networks and wireless networks in order to enhance the system throughput and robustness…. This paper exploits CUDA for network coding and homomorphic hashing…. The results show that for network encoding the CUDA approach shows an 80X speedup over same generation CPUs…." Authors: Xiaowen Chu‚ Kaiyong Zhao‚ Mea Wang; Hong Kong Baptist University; University of Calgary‚ Alberta‚ Canada. See:
Submit Your Work
Have a CUDA-related paper or research? Show it on CUDA Zone:
Zymeworks Inc. is a Canadian computational biotech company developing technology for optimizing protein therapeutics. Zymeworks is looking for a talented software engineer to take part in developing high performance molecular simulation and analysis tools on GPUs. See posting here: See more CUDA and GPU computing-related job postings here:
CUDA Education
PGI CUDA Fortran Webinar
PGI Fortran from the Portland Group features GPU acceleration. A live webinar will be presented by PGI to provide an intro to the product and its GPU acceleration capabilities. The webinar will take place on Wednesday‚ March 24‚ at 9:00 a.m. Pacific time. Register now:
GPU Computing Webinars (CUDA C and OpenCL)
Current schedule:
Acceleware-Certified CUDA Training
Silicon Valley‚ March 24-25:
Calgary‚ April 12-16:
CUDA and GPU Computing Courses
Over 310 universities are teaching CUDA and GPU Computing courses. See the list:
Call for Papers
Symposium on Chemical Computations on GPGPUs. Abstracts due 4/5/10. See:
– CUDA Video on YouTube:
– CUDA Toolkit:
– Developer Guides:
– Programming Massively Parallel Processors, by D. Kirk, W. Hwu:
– Follow CUDA & GPU Computing on Twitter:
– Network with other developers:
– Stayed tuned to GPGPU news and events:
– Learn more about CUDA on CUDA Zone:
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.

2701 San Tomas Expressway, Santa Clara, CA 95050

Copyright © 2010 NVIDIA Corporation. All rights reserved.