AMD FireStream™ 9170: Industry's First GPU with Double-Precision Floating Point
AMD FireStream 9170 Specifications
- Powered by next-generation ATI GPU from AMD
- Parallel processing architecture with 320 stream cores
- Up to 500 GFLOPs single precision performance
- 2GB GDDR3 on-board memory
- Double Precision Floating Point
- PCIe 2.0 x16 interface
- < 150W power consumption
- Memory export
- BIOS settings optimized for stream processing
- API and OS Support
- Windows XP, XP64
- Linux 32 and Linux 64
System Requirements
- PCI Express® based server or workstation with available x16 lane graphics slot
- 500W or greater power supply
- 512MB of system memory
Optimized for Computation
- Double-precision floating point for scientific and engineering codes
- 2GB on-board memory
- 320 stream cores drive up to 500 GFLOPS single-precision performance
- Asynchronous DMA, allowing data transfers without interrupting streams processor or CPU
Accelerating data-parallel algorithms in a variety of applications
- Seismic processing
- Financial analysis
- Engineering analysis
- Rendering
- Computational chemistry
- Genetic research
- Physics
- Video editing
- Security
AMD Warranty and Support
- Enterprise-level support
- Three year limited product repair/replacement warranty
- Toll-free phone and email access to technical support team
AMD FireStream stream processors deliver the tremendous parallel processing power of the GPU to compute-intensive applications for scientists, engineers and consumers. Using this leading-edge technology, developers can dramatically increase the performance of their critical algorithms, allowing users to get faster results or work on larger problems.
AMD leverages its expertise in CPUs, GPUs and software to deliver stream processing hardware solutions along with a comprehensive software development stack. Fortune 1000 companies, leading software developers, and academic institutions are demonstrating today that stream computing is the key to tremendous performance in a variety of applications.
The AMD FireStream 9170 provides the industry's first double-precision floating point capability on a GPU. With 2GB GDDR3 memory on board and single-precision performance of up to 500 GFLOPS, it is ideal for the most demanding compute-intensive, data-parallel tasks.
Using ground-breaking 55nm process technology, even this large-memory board consumes less than 150W, making it easy to deploy in a variety of server and workstation packages.
AMD FireStream SDK: Open systems approach drives adoption
The AMD FireStream SDK delivers all the tools developers need to create and optimize applications on stream processors. Developers can begin with Brook+, an AMD-enhanced and supported implementation of Brook, the popular open-source C-level language and compiler.
Math functions can be implemented using a new release of AMD Core Math Library (ACML), which accelerates these functions on the GPU, and tools like ATI Shader Analyzer and AMD CodeAnalyst help identify and correct performance issues.
The 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 APIs and specifications, enabling performance tuning at the lowest level and development of third party tools.
| Features | Benefits |
|---|
| Up to 500 GFLOPS of compute power | Dramatic acceleration for critical algorithms |
| Double-precision floating point | Appropriate for scientific and engineering calculations |
| Asynchronous DMA | Move data without interrupting streams processor or CPU |
| 2GB on-board memory | Compute on large datasets without CPU traffic |
| Memory export | Provides flexible storage for intermediate results |
| <150W power consumption; 55nm process | Optimal performance-per-watt |
|
