Sapphire today announced next-generation AMD FireStream 9250™ processor accelerates scientific and engineering calculations, efficiently delivering supercomputer performance at up to eight gigaflops-per-watt. It features 1GB of GDDR3 memory. The FireStream™ 9250 and the supporting SDK are available in beginning of November 2008 while AMD FireStream™ 9170 is currently available for purchase now.

Sapphire Technology today announced next-generation AMD FireStream 9250™ processor accelerates scientific and engineering calculations, efficiently delivering supercomputer performance at up to eight gigaflops-per-watt —

AMD’s next-generation stream processor, the AMD FireStream™ 9250 is specifically designed to accelerate critical algorithms in high-performance computing (HPC), mainstream and consumer applications. Leveraging the GPU design expertise of AMD’s Graphics Product Group, AMD FireStream™ 9250 occupies a single PCI-E 2.0 slot only, breaks the one teraflop barrier for single precision performance. AMD FireStream™ 9250 can deliver performance of up to eight GigaFLOPS per watt in single-precision, and more than 200 gigaflops raw double-precision performance.

The AMD FireStream™ 9250’s compact size makes it ideal for small 1U servers as well as most desktop systems, workstations, and larger servers and it features 1GB of GDDR3 memory, enabling developers to handle large, complex problems. For example, developers are reporting up to a 55x performance increase on financial analysis codes as compared to processing on the CPU alone, which supports their efforts to make better and faster decisions.*

AMD Stream SDK leverage open source technology and software technology partners to provide a robust, open, multi-layer development environment on 32 and 64 bit Linux and Windows environments.

Developers can begin with Brook?, the popular open-source C-level language and compiler. Math function can be implemented using a new release of AMD Core Math Library for the GPU (ACML-GPU). Tools like GPU Shader Analyzer and AMD Code Analyst help identify and correct performance issues.

AMD’s Compute Abstraction Layer (CAL) provides low level access to the GPU for development and performance tuning. AMD’s open systems approach allows developers access to all key API’s and specification, enabling performance tuning at the lowest level and development of third party tools.

How AMD Stream Computing helps
AMD Stream Computing provides industry-leading performance to handle the hardest HPC Application.

  1. Up to 8 GFLOPs per watt—One teraflop of raw performance at under 150 watts.
  2. Commodity GPU technology—Affordable 1 TeraFLOP performance
  3. Multiple software development paths from high level languages and libraries to low level inter-block communications.
  4. Solution scalability through combinations of CPUs and GPUs to match specific problem needs.
  5. Support from AMD engineers and partners throughout your development process.

Industries and Applications
Financial Analytics
Energy, Oil and Gas
Life Sciences
Computer Aided Engineering (CFD, FEA, etc.)
Consumer

Availability
The FireStream™ 9250 and the supporting SDK are available in beginning of November 2008. AMD FireStream™ 9170, the industry’s first double-precision floating point stream processor, is currently available for purchase.