A couple of days ago, NVIDIA announced the Quadro FX 5800 GPU which comes equipped with a massive 4GB of RAM and 240 CUDA processors. In other words, if you have some serious number crunching to do (and most scientists are in such a predicament) and about $3500 to spend, then you are one lucky person.
According to the Press Release,
The Quadro FX 5800 graphics card offers unprecedented performance and scalability to rapidly visualize and interpret massive datasets that until now were unattainable on a workstation graphics board. Offering up to 240 CUDA(TM) programmable parallel cores and the industries first 4GB of graphics memory, the Quadro FX 5800 graphics card is ideally suited for oil and gas exploration, medical imaging, styling and design, and scientific visualization.
In the past, I have more than once expressed my fondness for this type of hardware and its usefulness in artificial intelligence and robotics applications. Other than NVIDIA, ATI also make programmable general purpose GPU hardware and, of course, Intel are also developing multi-core CPUs with more than a handful of cores. Finally, don’t forget about the MIT startup Tilera Corporation and their Tile64 processor.
The next couple of years are going to be very exciting with scientists and engineers trying to solve larger and larger problems using this and the next generation of massively parallel processing systems. I can’t wait!