|
CUDA SPOTLIGHT |
GPU-Accelerated Medical Image Processing
For this week’s Spotlight we interviewed Anders Eklund, a Ph.D. student at Linkoping University in Sweden. Anders is affiliated with the University’s Center for Medical Image Science and Visualization (CMIV). Here’s a preview of our conversation: |
|
|
NVIDIA: Anders, what is your research focused on? |
Anders: I work with algorithms for medical image processing, such as image registration and image denoising. My research is especially focused on functional magnetic resonance imaging (fMRI), where you try to find brain activity from magnetic resonance images (MRIs) of the brain.
My interests include brain-computer interfaces (BCI) with real-time fMRI, where the fMRI data is processed in real-time as the subject is in the MR scanner. A brain-computer interface could help people communicate who are paralyzed or suffer from Locked-in syndrome. |
NVIDIA: How does GPU computing play a role in your work? |
Anders: GPU computing is very important for me and the research group in which I work, as many of the algorithms that we develop are very computationally demanding. To be able to develop, evaluate and improve an algorithm, it really helps if the processing time for one run can be reduced from minutes to seconds, or from hours to minutes. |
NVIDIA: As computing becomes more powerful, what can we look forward to? |
Anders: One exciting trend within medical imaging is to move the processing of the data into the surgery room, such that the medical doctors can get real-time feedback during surgery. |
|
- Read the full interview with Anders Eklund
- See video on 4D Image Denoising
- See video on the Brain-Computer Interface
Editor’s Note: Anders will present on 4D medical image processing at GTC 2012 in May in San Jose, Calif.
(To suggest a CUDA Spotlight, email cuda_week_in_review@nvidia.com)
|
|
CUDA DEVELOPER NEWS |
back to the top |
New CUDA Book |
Congrats to author Rob Farber on the release of "CUDA Application Design and Development." Published by Morgan Kaufmann and available on Amazon, this book is designed for software developers who want to leverage GPU programming with CUDA to increase efficiency.
- See: http://amzn.to/rQZrqS
SC11 News Wrap–Up
This year’s Supercomputing conference (SC11) in Seattle was action-packed. Here is a roundup of some of the news highlights:
- OpenACC Programming Standard for Parallel Computing Unveiled
- BSC to Deploy World’s First ARM-Based CPU/GPU Hybrid Supercomputer
- NVIDIA Tesla GPUs to Accelerate NCSA Blue Waters Supercomputer
Accelerate Apps Easily with Directives on GPUs
As part of the recently announced "2X in 4 Weeks" program, Professor Amin Kayali from the University of Houston accelerated his micromagnetic simulation code by 20X in less than two days by inserting directives, or "compiler hints," into his CPU code. The compiler uses these directives to map compute-intensive portions of the code to the GPU. Want to accelerate your code? To help you get started, NVIDIA and PGI are offering a free 30-day license of the directives-based PGI Accelerator compiler.
- See: http://www.nvidia.com/object/tesla-2x-4weeks-guaranteed.html?cid=dev#19
GTC Asia Just Around the Corner
The next GPU Technology Conference (GTC) will be held in Beijing on Dec. 14-15. The event will be kicked off with a keynote by NVIDIA CEO Jen-Hsun Huang, followed by a great line-up of speakers from the Chinese Academy of Sciences, Harvard, HP Labs, Tokyo Institute of Technology, Tsinghua University and other leading organizations.
- See: http://www.gputechconf.cn/page/home-en.html
Help Name the New 'CUDA on ARM Dev Kit'
At SC11 we announced plans for an ARM-based GPU computing development kit to support the growing demand for energy-efficient HPC initiatives. The kit will feature a quad-core NVIDIA Tegra 3 ARM CPU accelerated by a discrete NVIDIA GPU. We are very excited about this technology and are looking for the perfect codename. Call up your creative juices and submit an idea. If your suggestion is chosen, you will receive a free DevKit when it launches next year!
- See: http://bit.ly/cudaarmnews
|
|
NEWS FROM ACADEMIA
|
back to the top
|
Map the World of GPU Computing |
If your university or research organization has a GPU compute cluster, let the world know about it by adding it to the new GPU Computing map. For info, read the blog post by NVIDIA’s Devang Sachdev.
- See: http://bit.ly/sE86d3
New CCOEs
Two new institutions have been named CUDA Centers of Excellence: The Barcelona Supercomputing Center and Lomonosov Moscow State University. The CUDA Center of Excellence designation is the highest honor given to institutions for ground-breaking work leveraging NVIDIA GPUs and CUDA.
- See: http://bit.ly/vAhZew
|
|
|
NEW ON THE BLOG |
back to the top |
|
Tokyo Tech Nabs Gordon Bell Prize, by Sumit Gupta
Exascale: An Innovator's Dilemma, by Andy Walsh
GPUs Give Students Edge in SC11 Cluster Competition, by Andy Walsh |
|
REPLAY OF THE WEEK |
back to the top |
|
NEW: Our pick for this week is: Why the Future of HPC Will Be Green (SC11) by Steve Scott, NVIDIA
- See: http://nvidia.fullviewmedia.com/fb/nv-sc11/tabscontent/archive/313-wed-scott.html |
|
CUDA JOBS |
back to the top |
|
NEW: Cal Tech is seeking a Computational Scientist to engineer scientific software to make effective use of accelerators, particularly GPUs. Initial responsibility will be optimizing codes to exploit a large new hybrid (CPU/GPU) cluster in the Division of Geological and Planetary Sciences. Apps include Bayesian models of fault slip during large earthquakes, inverse models of the Earth’s interior structure, large-scale remotely sensed image processing and models for use in rapid tsunami early warning systems.
- See: https://jobs.caltech.edu/applicants/jsp/shared/Welcome_css.jsp (Note: Search on ‘computational scientist’)
(To submit a job listing, email cuda_week_in_review@nvidia.com)
|
|
GPU MEETUPS |
back to the top |
|
The GPU Meetups offer a great way to learn about GPU computing and meet interesting people in a relaxed environment:
Silicon Valley GPU Meetup, Mon., Dec. 5, 6:15 pm |
New York GPU Meetup, Thurs., Dec. 8, 6:00 pm (Special topic: The Business of GPUs) |
Brisbane GPU Meetup, Australia, Thurs., Dec. 15, 6:00 pm |
(Would you like to start a Meetup? Email cuda_week_in_review@nvidia.com)
|
|
|
CALLS FOR PAPERS AND POSTERS |
back to the top |
|
GTC U.S. 2012 (May 14-17)
Poster deadline: Dec. 8
http://www.gputechconf.com/page/participate.html
https://gtc-submissions.confreg.com/
|
|
CUDA Calendar |
back to the top
|
- AGU (American Geophysical Union) Meeting
- Intro to GPU Programming Workshop — La Maison de la Simulation
- NEW: CUDA Training by Acceleware with Microsoft
- NEW: CUDA Training by T- Platforms and Moscow State University
- GTC Asia
- LibJacket CUDA Library for Maximus — AccelerEyes (Webinar)
Dec. 15, 2011 |
Learn to integrate computations with visualizations in a CUDA-based app through simple visualization functions for plotting, image and volume rendering, and more. |
http://bit.ly/rdZ8pHs |
2012
- CUDA Programming 1-Day Course - Delft University of Technology
- NEW: PRACE Winter School at CINECA
(To list an event, email: cuda_week_in_review@nvidia.com) |
|
|
CUDA RESOURCES |
back to the top |
|
|
– Sign up: www.nvidia.com/paralleldeveloper |
|
– List of CUDA-enabled GPUs: http://developer.nvidia.com/cuda-gpus |
|
– 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
– Stay 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 |
|
– 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 |
|
– SC11 presentations: http://www.gputechconf.com/page/gtc-on-demand.html |
– The Third Pillar of Science: www.nvidia.com/object/race-for-better-science.html |
– GTC 2010 presentations: www.nvidia.com/gtc |
|
ABOUT CUDA |
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 as well as APIs such as OpenCL and DirectCompute. Send comments and suggestions to: cuda_week_in_review@nvidia.com
|
|
Click here to opt in specifically to CUDA: Week in Review. |
|
Copyright © 2011 NVIDIA Corporation. All rights reserved. 2701 San Tomas Expressway, Santa Clara, CA 95050. |