AMD FirePro™ S9170 Server GPU
The Best GPU for Compute Just Got Better
Key Features:
• Passively cooled solution for server environments
• AMD Graphics Core Next architecture
• 2,816 stream processors (44 compute units)
• 5.24 TFLOPS peak single-precision floating point
• 2.62 TFLOPS peak double-precision floating point
• Full throughput double-precision
• Error-correcting code memory support (external only)
• 32GB ultrafast GDDR5 memory
• 512-bit memory interface
• Up to 320GB/s memory bandwidth
• 275W maximum power consumption
• Support for SMBus temperature reporting at boot-up
• AMD PowerTune technology
• AMD STREAM technology
• OpenCL™, OpenGL support
• PCIe® x16 bus interface, PCIe 3.0 compliant
• Full-height/full-length dual-slot form factor
• Headless display support
• Linux OS support (64- and 32-bit)
3
4
Accelerate your most complex workloads in scientific computing, data
analytics, or seismic processing, and witness the power of the world’s
first 32GB server GPU, the AMD FirePro™ S9170. With industry-leading
32GB of GDDR5 memory and up to 2.62 TFLOPS of peak double-precision
performance, the choice is clear.
Unparalleled GPU Compute Performance
The AMD FirePro™ S9170 features full throughput double-precision, enabling the
card to achieve up to 2.62 TFLOPS of peak double-precision compute, making the
AMD FirePro S9170 the fastest single-GPU server card available today. The AMD
FirePro S9170 delivers up to 40% more performance in DGEMM heavy applications
than the Tesla K80 dual-GPU card.
Industry-leading Memory Configuration
The AMD FirePro S9170 is the industry’s first server GPU with 32GB ultra-fast
GDDR5 onboard memory — 2.7x the memory of the competition.2 It also features
a 512-bit memory interface for up to 320GB/s of memory bandwidth, helping to
improve overall workload speed and system responsiveness.
1
• FCC, CE, C-Tick, BSMI, KCC, UL,VCCI, RoHS, and WEEE compliance
• Designed, built, and tested by AMD
• Planned minimum three-year life cycle
• Limited three-year warranty
Future Ready
AMD is a firm believer and supporter of open standards such as OpenCL™, OpenGL,
OpenMP™, and OpenACC™ and supports these standards on the new AMD FirePro
S9170. We believe most people in the HPC community want open standards as
the de facto way of running their projects and simulations, and AMD is committed
to supporting this goal, working extensively with partners within the community
to drive open standards forward.
AMD FirePro™ S9170 Server GPU
Features Benefits
AMD Graphics Core Next (GCN)
Architecture
5.24 TFLOPS of Peak Single-Precision
Floating Point Performance
2.62 TFLOPS of Peak Double-Precision
Floating Point Performance
Full Throughput Double-Precision
Error Correcting Code (ECC) Memory
Support
Multi-GPU Support
32GB GDDR5 Memory
AMD PowerTune Technology
3
The first GPU architecture designed with compute workloads in mind. Engineered for high
utilization, high throughput, and multitasking.
Helps speed up time required to complete single-precision operations used within Video Eects
and Rendering, Signal Processing, Transcoding, and Digital Rendering applications where high
performance takes precedence over accuracy.
Helps speed up time required to complete double-precision operations used within Computational
Fluid Dynamics, Structural Mechanics, Reservoir Simulation, and Aerodynamics applications, where
numerical precision is mission critical.
AMD FirePro S9170 server GPUs feature full throughput double-precision. Unlike competing server
cards that are not optimized for double-precision, AMD FirePro S9170 cards oer the most doubleprecision performance1, completing compute-intensive tasks faster than ever before.
Helps ensure the accuracy of computations by correcting any single or double bit error as a result of
naturally occurring background radiation. External support only.
Combine multiple AMD FirePro S9170 server GPUs in a single system and leverage the combined
processing power to tackle the most compute-intensive workloads.
Equipped with 32GB of GDDR5 memory, the FirePro S9170 can accelerate memory-intensive
applications and process larger and more computationally complex workflows with ease.
AMD PowerTune Technology is an intelligent power management system that monitors both GPU
activity and power draw. AMD PowerTune optimizes the GPU to deliver low power draw when GPU
workloads do not demand full activity, and delivers the optimal clock speed to ensure the highest
possible performance within the GPU’s power budget for high intensity workloads.
AMD STREAM Technology
4
intensive workflows leveraging the massively parallel processing power of AMD GPUs, and
accelerates many applications beyond just graphics.
Powers the ecosystem that enables AMD FirePro™ S-Series server cards to be used for compute-
OpenCL™ 2.0 Support
Hardware compatible with other AMD
FirePro™ products
The AMD FirePro S9170 server GPU supports OpenCL™ 2.0, allowing developers to take advantage
of new features that give GPUs more freedom to do the work they are designed to do.
The AMD FirePro™ S9170 board dimensions are identical to the AMD FirePro™ S9150 and AMD
FirePro™ S9100. This allows for support of custom brackets and retention mechanisms designed for
those other boards to be used on the new AMD FirePro S9170.
For more information, please visit www.amd.com/firepro
1. AMD FirePro™ S9170 graphics delivers up to 2.62 TFLOPS peak double-precision floating point performance. Nvidia’s highest performing single-GPU server card in the market as of May 2015 is the Tesla K40, with
up to 1.43 TFLOPS peak double-precision performance. Nvidia’s highest performing dual-GPU server card in the market as of May 2015 is the Tesla K80, with up to 1.87 TFLOPS of peak double-precision performance.
Visit http://www.nvidia.com/object/tesla-servers.h™l for Nvidia product specs. FP-131
2. AMD FirePro™ S9170 graphics is equipped with 32 GB GDDR5 memory, while Nvidia’s K40 offers 12GB GDDR5, and Nvidia’s dual GPU K80 offers 24GB GDDR5 (12GB per GPU).
Visit http://www.nvidia.com/object/tesla-servers.h™l for Nvidia product specs. FP-135
3. AMD PowerTune and AMD ZeroCore Power are technologies offered by certain AMD FirePro™ products, which are designed to intelligently manage GPU power consumption in response to certain GPU load conditions. Not all
products feature all technologies – check with your component or system manufacturer for specific model capabilities.
4. AMD STREAM technology is a set of features offered with select AMD FirePro graphics cards for the acceleration of compute-intensive workflows. Not all products have all features, and full enablement of some capabilities may
require complementary software. Check with your system manufacturer for specific capabilities and supported technologies.
© Copyright 2015 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, FirePro, and combinations thereof are trademarks of Advanced Micro Devices, Inc. OpenCL and the OpenCL logo are trademarks of Apple
Inc. used by permission by Khronos. Other names are for informational purposes only and may be trademarks of their respective owners. PID# 156778-A