Stream processing is a relatively new concept in parallel processing that allows for considerable performance boosts under certain scenarios, especially where applications depend heavily on calculations. AMD has pursued stream processing before and has announced today their new thin hardware interface called "Close To Metal" (CTM). According to the company, CTM gives developers access to native instructions set and memory "of the massively parallel computational elements in AMD Stream Processors."
While we have come to expect mostly CPU technologies from the Sunnyvale manufacturer, this product actually falls in a different category touted as General-Purpose Computing on Graphics Processing Units. So in reality this AMD Stream Processor board is a PCI-Express card using the ATI R580 GPU (the one used in the ATI Radeon X1900 graphics cards), packed with 1 GB of GDDR3 memory and a modified memory controller that allows for stream computing applications. AMD would seem to be aiming at the supercomputer crowd with the "Stream Processor" for which it plans to charge $2600 per board.
Using CTM today, AMD is working with a number of companies to deliver the tools ecosystem for stream computing, said Marty Seyer, senior vice president, Computational Product Group, AMD. As part of our Torrenza initiative and CTM, AMD is enabling companies to work with best-of-breed vendors that understand how to optimize their software across all processor architectures, whether in stream processors or high-performance CPUs. For these organizations, the development of highly capable, and efficient software is their business, not a sideline. Allowing open innovation to flourish will ultimately enable better software, with more features to come to market faster than any proprietary approach.