Harness your GPU for Folding@Home and other applications

ATI today announced their new "Stream Computing" initiative, which takes the GPU capabilities and ambitiously puts its horsepower towards applications aside from graphics. ATI and NVIDIA already demonstrated that physics calculations are perfect computationally expensive processes to run on GPUs with a huge improvement over traditional CPUs.  Today ATI is taking this a step further and putting their GPUs to tasks that I think many gamers and enthusiasts would not have expected just a few years ago: the sciences, commercial sector, and business sector can all benefit Stream Computing.  Most importantly, this opens up all sorts of new implications.  We already highlighted the difference in raw Gigaflops calculation power of a modern high end GPU versus a high end CPU - Stream Computing takes advantage of this differential, and the multiple cores of GPUs to usher a different application paradigm.  The graphics companies have been talking about it for a while: parallelism gives GPUs far superiority to their CPU counterparts in graphics, but also in many other areas.

In particular interest to enthusiasts and 'humanity', ATI is very excited that Stream Computing will enable the next generation Folding@Home clients to harness the power of an X1900 class GPU to accelerate Folding@Home operations as much as 20-40 times!  Yesterday, in a private conference call with media, Vijay Pande, associate professor of chemistry, and director of the Folding@home distributed computing project for Stanford University illustrated the drastic possibilities this opens up: imagine 3 years worth of protein folding done in the span of 1 month!  The implications of this type of technology, and its impact as GPUs become ever more powerful, are staggering.  Here are some of the examples ATI provides of Stream Computing's potential:

ATI's high-end processor today makes use of 48 compute cores that results in an order of magnitude processing speed-up. In certain applications(3), ATI processors perform up to 40 times faster than competing processors. The accelerated processing associated with stream computing has implications for a number of fields now and in the future, as the ecosystem around stream computing matures:

 

Scientific research - Today ATI's stream computing efforts are helping to save lives by driving life sciences to produce results faster in areas such as disease research, giving organizations the option to do more granular studies in the same amount of time as in the past(4). ATI announced today that Stanford University will make available a new distributed computing application that takes advantage of ATI processors for disease research. In the future, climate research may also benefit from stream computing as analysis of large data sets for storm and hurricane forecasting can be done faster or in more detail, potentially resulting in the issuing of warnings longer in advance of severe weather, and ultimately a better understanding of the world's climate.

 

Homeland security - Communications analysis and facial recognition can be drastically improved using stream computing, with implications for airport security, as well as photograph and video analysis.

 

Financial forecasting - Major institutions have been using server farms to do risk assessment using Monte Carlo simulations, and for derivatives pricing using models like Black-Scholes. Simulations conducted by PeakStream, Inc. using ATI hardware shows that stream computing can provide these companies with more detailed answers in significantly less time, letting them make the business decisions they need to faster, and giving them a leg up on their competition.

 

Oil and gas - Companies are using stream computing to analyze more data in shorter periods of time to more quickly and reliably discover where resources lie, speeding discoveries of crude oil deposits. ATI graphic processors in concert with PeakStream's software platform are allowing oil and gas companies to achieve significantly faster seismic data modeling.

 

Database searching - For search companies with incredibly large databases to organize and sort through, stream computing may offer a compelling business case providing increased processing power in less space.

 

Consumer applications - Software used by millions of people around the world, such as operating systems, office applications, and graphics applications, can benefit from stream computing. Any graphics-laden software that requires heavy processing can be accelerated.

 

Videogames - An area where ATI processors are already heavily used for graphics purposes, working with Havok, premier provider of software and services to interactive digital media creators, stream computing is resulting in life-like modeling of hair, cloth, smoke, liquid, and the physics behind them, giving gamers the most immersive experience possible.

 

Other areas that stream computing has the potential to impact in the future include enterprise software, product design and manufacturing, and digital media encoding among others.