CUDA: Week in Review
Tues., Dec. 20, 2010, Issue #44 - Newsletter Home
Welcome to CUDA: Week in Review, an online news summary for the worldwide CUDA and GPU computing community.
Editor's note: Happy holidays to all! Publication will resume in January 2011.
Developing Robots with CUDA
PhD student Martin Peniak is creating robots that can develop cognitive capabilities. We first learned about Martin when he wrote us a note on our Facebook page. Below is our interview with him:
NVIDIA: Martin, how did you start working with robots?
Martin: Humanoid robots and technology in general have always fascinated me. As I grew up, this curiosity broadened. I graduated from an engineering school in my home country, Slovakia, and then moved to the United Kingdom where I studied computing and astronomy. Now, I am doing a PhD at the University of Plymouth related to the iTalk project (Integration of Action and Language in Humanoid Robots).
NVIDIA: What is iTalk?
Martin: iTalk aims to create biologically-inspired artificial systems that can progressively develop cognitive capabilities through interaction with their environments. The project is coordinated by my supervisor, Professor Angelo Cangelosi.
NVIDIA: Tell us a bit about your current work.
Martin: Based on insights from neuroscience, psychology, robotics, linguistics and other domains, we argue that cognitive skills have their foundations in the morphology and material properties of our bodies. The iTalk project uses one of the most complex humanoid robots in the world. This intricate robotic platform - called iCub - is approximately 105 centimeters tall and weighs around 20 kilograms. It was designed by the RobotCub Consortium.
NVIDIA: How are you using CUDA and GPU computing?
Martin: For my research, I use CUDA-enabled software called Aquila to develop complex artificial neural networks - inspired by those found in the brain - for the real-time control of the iCub robot.
NVIDIA: How did you hear about CUDA?
Martin: I originally heard about CUDA from a Russian friend. Then, one day I found an article about a neural network implementation using CUDA and was impressed by the performance increases that were achieved. I showed the article to my colleague and after many discussions we agreed that GPU processing was exactly what we needed in our lab, as most of the systems can easily be parallelized. We looked into OpenCL as an alternative but the CUDA framework provided much more support and the API was really good. We ordered six servers with Tesla C1060 and GeForce GTX470 cards and created a Linux-based supercomputing cluster capable of performing over 12 TFLOPS (trillion floating point operations per second).
NVIDIA: Why do you need the power of GPU computing?
Martin: The artificial neural networks often consist of thousands of neurons that are connected to many other neurons through millions of synaptic connections. The multidimensional input from various senses is abstracted into internal representations. This is achieved through self-organizing maps which resemble the cortices in the brain. Often reaching sizes of several thousand neurons, these maps are abstracting the original visual data by applying filters to millions of pixels. Apart from this visual processing, the system must also work with linguistic and somatosensory inputs while performing millions of calculations needed to activate the neural network every 50-100 milliseconds.
NVIDIA: How are the results so far?
Martin: The CUDA framework accelerated the online neural network control several hundred times on average, and the algorithms responsible for iCub’s training showed around a 50X speed increase. I have developed both CPU and GPU versions and although I haven’t completed extensive optimizations, the nice thing about CUDA is that simply by naive parallelization of the CPU code one can achieve massive speedups using GPU devices. As quantum computing is still in its infancy, to me it seems that massively parallel GPU processing is the way to move forward since CPU architectures are simply not suited for parallel tasks. They consume too much energy and do not scale well.

  - For more info, see the YouTube video here and the NVIDIA blog post here.

  Would you like to be featured in the CUDA Spotlight? Email us at
Business Intelligence in the Cloud
Jedox, supplier of business intelligence solutions, announced that its Palo GPU technology is available via Amazon Elastic Compute Cloud (EC2). Kristian Raue of Jedox commented: "We have reached a major milestone for future development with GPU technology."
  - For info on Jedox, see:
  - For info on Amazon EC2 (Cluster GPU Instances), see:

New Molecular Dynamics Code with CUDA Support
DL_POLY is a general purpose molecular dynamics (MD) simulation software developed at Daresbury Laboratory in the U.K. by Dr. Ilian Todorov and Dr. William Smith. The newest release (DL_POLY_4) was developed in collaboration with the Irish Centre for High-End Computing to harness the power of CUDA and NVIDIA GPUs. DL_POLY_4 is free of cost to academics. Commercial organizations may contact Dr. Todorov at
  - See:
NEW: Each week we will highlight a session from GTC 2010. Here’s our pick for this week:
    Power Management Techniques for Heterogeneous Exascale Computing
    Xiaohui (Sean) Cui - Oak Ridge National Laboratory (40 mins.)

University of Delaware, Global Computing Lab: Graduate research assistant to work on Monte Carlo methods accelerated by GPUs. Ideal candidate has C/C++ skills; CUDA/OpenCL skills; MPI/OpenMP skills; and exposure to parallel performance optimization and profiling. Contact:
Oak Ridge National Laboratory: Post-doc research associate in computational statistics, in area of climate data analysis. PhD in statistics or C.S. preferred with strong interest in parallel computing. Funded by three-year project called "Visual Data Exploration and Analysis of Ultra-Large Climate Data" (Dept. of Energy).
December 2010

Tutorials on GPU Programming - HiPC 2010
Dec. 19-22, Goa, India

NEW: International Parallel & Distributed Processing Symposium - IEEE
Call for papers: Dec. 22, 2010 (Parallel Computing & Optimization Workshop)
Event: May 16-20, 2011, Anchorage

January 2011

Scientific Computing in the Americas: The Challenge of Massive Parallelism
Jan. 3-14, 2011, Valparaiso, Chile

NEW: GPU & Parallel Computing Workshop - SagivTech and Microsoft R&D
Jan. 6, 2011, Herzliya, Israel (at Microsoft R&D)
Note: Free of charge, pre-registration required
For more info, contact:

NEW: CUDA/OpenCL Training - Acceleware and Colfax
Jan. 17-21, 2011, Sunnyvale, Calif.
For more info, contact:

NEW: Optimizing Financial Modeling/Chicago - Wolfram Research
Jan. 25, 2011, Chicago
Featured Speaker: Dr. Michael Kelly

NEW: Optimizing Financial Modeling/New York - Wolfram Research
Jan. 27, 2011, New York
Featured Speaker: Dr. Michael Kelly

February - December 2011

NEW: GPU Computing Session, German Physical Society Conference
March 13-18, 2011, Dresden, Germany

NEW: ASIM Workshop 2011 - ASIM and Technische Universitat Munchen (TUM)
March 14-16, 2011, Leibniz, Germany
Theme: Trends in Computational Science & Engineering: Foundations of Modeling & Simulation

NEW: Application Accelerators in High Performance Computing (SAAHPC 2011)
Call for papers: May 6, 2011
Event: July 19-21, 2011, Univ. of Tennessee, Knoxville, Tennessee

Intelligent Vehicles Conference - IEEE
June 5-9, 2011, Baden-Baden, Germany

Internat'l. Conference on Computer Systems and Applications
June 27-30, 2011, Sharm El-Sheikh, Egypt

– CUDA Training from EMPhotonics:
– CUDA Training from Acceleware:
– CUDA Certification:
– GPU Computing Webinars:

(To list an event, email:

Video Recommendation
– The Third Pillar of Science:
GPU Technology Conference
– Presentations and keynotes from GTC 2010:
– List of CUDA-enabled GPUs:
CUDA GPU Computing Forum
– Link to forum:
CUDA and Parallel Nsight Overview
– Blog post and video:
CUDA Downloads
– Download CUDA Toolkit 3.2:
– Download OpenCL v1.1 pre-release drivers and SDK code samples (Log in or
   apply for an account
– Get developer guides and docs:
CUDA and Academia
– Learn more at
CUDA on the Web
– See previous issues of CUDA: Week in Review:
– 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 Recommended Reading
– Read Kudos for CUDA:
– Read Supercomputing for the Masses, Part 20:
– Read CUDA books:
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. Send comments and suggestions to:
Stay in Touch with NVIDIA
Twitter   Follow GPU Computing on Twitter
Facebook   Become a fan of NVIDIA on Facebook
NVIDIA online profiles   See list of NVIDIA online profiles

Click here to opt in specifically to CUDA: Week in Review.

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