Friday, 11 June 2010



The World Is Parallel: Teaching CUDA

Posted: 11 Jun 2010 10:59 AM PDT

High-performance programming on GPUs has broad potential in a wide array of fields, from chemical research to image processing to business analytics. But bringing these advantages to potential beneficiaries in fields other than computer science requires the creation of applications to use the power of parallel processing. And creating those programs requires educating a new generation of software developers in the techniques of general purpose computing on GPUs. One way that NVIDIA is doing its bithelps this process along is by sponsoring graduate fellowships to students working on Ph.Ds in computer science or allied fields "who are researching topics that will lead to major advances in a number of fields, and are investigating innovative ways of leveraging the power of the GPU."


Anjul Patney and Duane Merrill are two of this year's NVIDIA graduate fellows. Patney. Patney is a third-year doctoral student at the University of California at Davis, working under John Owens, a specialist in general purpose programming on GPUs. "I have a strong interest in understanding the evolution of programmable graphics pipelines, a possibility recently enabled by the flexible CUDA, OpenCL and Direct Compute programming models.." Patney says. "I am interested in identifying both software and hardware principles that will define these pipelines in the future. In this pursuit I often run into some very intriguing problems in graphics and parallel computing. It's a lot of fun."

His area of interest is graphics rendering. As he wrote in his research statement in applying for the fellowship: "My research goal is to explore infrastructures that allow complex, dynamic, yet fully customizable data structures and algorithms to help build new rendering schemes. I wish to deploy them in an open-source library, to serve as an abstraction over which flexible graphics applications can efficiently interact with the underlying hardware."

Patney was first exposed to parallel processing techniques as an undergraduate at the Indian Institute of Technology, Delhi. "Early training in parallel and multicore programming is very important for interested undergrads," he says.


Merrill is a doctoral student at the University of Virginia. Merrill is working on applying the techniques of GPU processing to some of the basic building blocks of software, things such as sorting, the efficient processing of lists, and graph problems that he describes as "the nuts and bolts of data-intensive computer science, yet are the hardest to make scream on the GPU."

One of the challenges of GPU programming is that since it is relatively new—the first version of CUDA was only released three years ago—programmers have not had time to create the big libraries of pre-tested, ready-to-run code that can make software development must faster and easier—a situation Merrill hopes to help remedy. "We simply don't have many collections of reusable software components at our disposal, particularly at [the level] where the brunt of cooperative performance gains are derived," he says. "This is the hardest software to construct, make fast, and get correct," he says. "This is why the prospect of writing extremely high performance, high quality code for GPUs is such a challenge."

Merrill did his undergraduate work at Virginia, then spent several years in industry before returning to graduate school. In his undergraduate days, GPU programming "wasn't on the radar screen," but that has changed. "The experience is much different these days," he says. "Many of the undergraduate computer science students in our department opt to take a parallel-computing elective that has units on SPMD [single-program, multiple data] computing. This course is usually instructed by my advisor, Andrew Grimshaw, and I teach the GPGPU unit, in which we use the CUDA framework exclusively."

In addition to a $25,000 stipendfor the 2010-11 academic year, the fellowship also provides engineering and technical support to 9 graduate students at 9 different universities. "The relationship NVIDIA fosters with university researchers through its fellowship program provides a conduit for ideas and technology to flow between academia and industry," says NVIDIA researcher and former NVIDIA fellow Jared Hoberock. "In addition to financial sponsorship, the fellowship affords student researchers a unique opportunity to cultivate a dialogue with the finest engineering minds in the industry ."

NVIDIA will be accepting applications for the 2011-12 graduate fellowships beginning in November. Details on the program and on application procedures are on the NVIDIA website.

NVIDIA 3D Vision Helps Send Cosmonauts to Mars

Posted: 10 Jun 2010 03:26 PM PDT

We have exciting news from Russia that NVIDIA 3D Vision is being used by the Russian federal space program to give cosmonauts a virtual reality experience of what it would be like to go on a mission to Mars.

The Mars-500 project involves more than 100 experiments to investigate the possibility of manned space flights to the Red Planet. But before actually sending cosmonauts into space, the program is conducting simulations that replicate the three parts of a mission: the journey to Mars, landing and exploration, and the flight back to Earth.

For the latest experiment, which began June 3, six cosmonauts were sealed in a hermetic environment that simulates the isolation, confinement and communication delays of interplanetary flight. They'll remain there for 520 days while observers study the physiological and psychological effects of the mission. For the experiment to be useful, the experience needs to be as convincing and immersive as possible – and that's where NVIDIA 3D Vision comes in.


NVIDIA has supplied the crew with 3D Vision laptop systems that they'll use primarily when they reach Mars and land on the planet. During their "flight" the crew will be able to use 3D Vision for entertainment (reportedly, one cosmonaut is an avowed gamer and was very happy to see these laptops). But it's during the 30-day operations phase that 3D Vision is most critical – that's when the crew will experience a simulated landing on Mars and get to explore the planet's surface.

NVIDIA's technology was chosen as the best solution for providing the most realistic, immersive 3D experience. Russian company JCSI, which is creating the interactive 3D model of Mars for the cosmonauts to explore, will simulate different environments – including dust storms, meteorites and craters – so the crew can experience a full range of highly realistic conditions as they navigate the planet's terrain.

Then it's another long journey back home. At least the in-flight movie will be in 3D.

For more information check out the Mars 500 Blog, Twitter profile and YouTube channel.

New Release of Flash 10.1 Turbocharges ION and GeForce PCs

Posted: 10 Jun 2010 02:36 PM PDT

Internet video just got a major speed boost. The new version of Adobe Flash Player 10.1 – launched to the general public today -- features GPU acceleration for PCs. That means you can enjoy your favorite videos from sites like, BBC iPlayer, or YouTube HD in full 1080p HD on an ION netbook, Optimus notebook, or GeForce-powered PC.

For years, Flash video ran only on the CPU, and if you didn't have the biggest and baddest processor, then you had to suffer through laggy, stuttering video. Those days are over. With your GPU now doing the video acceleration, you can enjoy crystal-clear 1080p HD video from around the 'Net.


Giving Back with the NVIDIA Foundation and GTC

Posted: 10 Jun 2010 02:11 PM PDT

We always strive to engage the community during NVIDIA corporate events and at this year's GPU Technology Conference (GTC) we continue that tradition by partnering with the NVIDIA Foundation — NVIDIA's philanthropic organization - to raise funds for education and cancer research.

During last year's inaugural GTC, we held a charity social in downtown San Jose. A percentage of food and drink revenues from the event were donated to fund a computer lab at McKinley Elementary School – the recipient of NVIDIAs Project Inspire. Project Inspire is an NVIDIA-hosted employee volunteer project where we spend an entire day beautifying a school in need. The GTC charity social resulted in the donation of 25 new computers.


This year, we aren't limiting our philanthropic/charity efforts to those folks who can join us in person. We'd like to invite anyone who is interested in participating, to make a $25 tax-deductible donation to one of the following two initiatives. All donations will be matched by the NVIDIA Foundation.

1. Compute the Cure. Providing cancer researchers with technology and financial support.
2. Visualize Math and Science. Using 3D simulations to get youth interested in math and science.

For more information or to make a donation, go here.

If you haven't done so already, check out the GTC event page and sign up here to follow GTC 2010 to get the latest updates.

No comments:

Post a Comment