AMD today launched the AMD FirePro S10000, the company's most
powerful server graphics card, designed for high-performance
computing (HPC) workloads and graphics intensive applications.
The AMD FirePro S10000 is the first professional-grade card to
exceed one teraFLOPS (TFLOPS) of double-precision floating-point
performance. It is also the first high-end card that brings a 5.91 TFLOPS of peak single-precision and 1.48 TFLOPS of double-precision floating-point calculations.
"The AMD FirePro S10000, equipped with our Graphics Core Next
Architecture, enables server graphics to play a dual role in
providing both compute and graphics horsepower simultaneously.
This is executed without compromising performance for users
while helping reduce the total cost of ownership for IT
managers," said David Cummings, senior director and general
manager, Professional Graphics, AMD.
The S10000 is a dual-GPU Tahiti card, packing a pair of 825MHz
Tahiti GPUs on a single board, with each GPU wired up to 3GB of
GDDR5 operating at 5GHz with a 384-bit interface. The launch of
the S10000 makes it the second dual-GPU server card to be
released this year, joining NVIDIA's K10.
| AMD FirePro S Series Specifications
|| AMD FirePro S10000
|| AMD FirePro S9000
|| AMD FirePro S7000
| Stream Processors
| Core Clock
| Memory Clock
|| 5.0GHz GDDR5
|| 5.5GHz GDDR5
|| 4.8GHz GDDR5
| Memory Bus Width
| Single Precision
|| 5.91 TFLOPS
|| 3.23 TFLOPS
|| 2.4 TFLOPS
| Double Precision
|| 1.48 TFLOPS (1/4)
|| 806 GFLOPS (1/4)
|| 152 GFLOPS (1/16)
| Transistor Count
| Manufacturing Process
|| TSMC 28nm
|| TSMC 28nm
|| TSMC 28nm
AMD's DirectGMA support removes CPU bandwidth and latency
bottlenecks, optimizing communication between both GPUs. This
also enables P2P data transfers between devices on the bus and
the GPU, bypassing any need to traverse the host's main memory,
utilize the CPU, or incur additional redundant transfers over
PCI Express, resulting in high throughput low-latency transfers.
Of course, the S10000 also supports OpenCL, the compute
programming language of choice among developers looking to take
full advantage of the combined parallel processing capabilities
of the FirePro S10000.
With the new card AMD is trying to hold off NVIDIA's Tesla and
VGX products, but the issue they face with the S10000 is one of
design and power consumption. Traditional high-end server cards
like the Nvidia Tesla K10
and FirePro S9000 are in the 225W-250W TDP range; meanwhile the S10000 significantly overshoots that,
coming in at 375W.
AMD's new server GPU will also face the competition of Nvidia's new K20, GPU, which has 5GB of memory and delivers 1.17 teraflops of double-precision performance and 3.52 teraflops of single-precision performance. Nvidia is also offering a faster K20X model, which has 6GB of memory and delivers 1.31 teraflops of double-precision performance. The K20X is three times faster than its predecessor, the Tesla M2090, which was released in the middle of last year.
AMD will be launching the S10000 at $3599, $1100 over the S9000.
AMD has already started shipping the card to key strategic