AMD has leapt into the teraflop computing arena with an accelerated Opteron dual-core processor based platform running a standard version of Microsoft Windows XP Professional.
The Opteron is accelerated with a couple of R600 Stream Processors, which will carry out a trillion floating-point calculations per second using a general multiply-add (MADD) calculation.
According to AMD, this represents a ten-fold performance increase over today's non-accelerated dual-core processor based servers.
AMD has made no secret of its plans to use co-processors interoperating with x86 microprocessors to provide supercomputer performance for specific applications.
Specifically, accelerated computing starts with an x86-based processor, and then adds an accelerator at the card level. AMD foresees integration in this space to start at the package level, moving to a completely integrated processor.
The Torrenza Stream Computing processor is AMD’s first effort in this area and is shipping now, he noted, which performs general-purpose computation on the graphics processing unit (GPU) and allows the floating point performance of the GPU to be an order of magnitude faster than the CPU. Changes to GPU architectures allowed this Polzin explained.
This has real commercial applications enabled by PeakStream software libraries, he said as he demonstrated an oil exploration application.
“Going forward, AMD will integrate at the device level with accelerated processors, and will integrate a CPU and GP called Fusion, focused on mobile applications,” said AMD senior fellow and chief platform architect Steve Polzin.
See Parallel Lines blog AMD more pragmatic than Intel with teraflop systems