AMD FirePro™ S10000 is a GPU compute power house, delivering record breaking single and double peak precision floating point performance. It’s the most powerful dual-GPU server graphics card ever created, delivering up to 1.3 times the single and up to 7.8 times the double peak precision floating point performance of the competition’s comparable dual-GPU product1.
Many High Performance Computing (HPC) implementations, like those focused on Genetic Sequencing, Structural Mechanics and Reservoir Simulation, are performing complex algorithms where the slightest miscalculation can significantly alter outcomes. As the world’s first professional grade graphics card to exceed one teraflop of peak double precision floating point performance - boasting an unprecedented 1.48 TFLOPS2 - the AMD FirePro™ S10000 is the right choice when precision matters.
For those implementations that don’t require the highest numerical accuracy, like Explicit Crash Test Simulations, Seismic Processing and Satellite Imaging, AMD FirePro™ S10000 graphics deliver an impressive 5.91 TFLOPS of peak single precision floating point performance. In fact, the AMD FirePro™ S10000 is 147% more powerful than the world’s fastest supercomputer in 1999. The ASCI Red was powered by more than 9,000 CPUs and achieved a peak speed of 2.38 single precision TFLOPS per second.
-Error correcting code (ECC) memory support to ensure computation precision and accuracy
-Capable of simultaneously processing both compute and graphics intensive applications, without requiring users to purchase a second graphics card
-Based on the world's first 28nm GPU architecture, AMD Graphics Core Next, designed for high utilization, high throughput and multitasking
Graphics Core Next (GCN) Architecture
The world's first 28nm GPU architecture, GCN is designed for high utilization, high throughput and multitasking. It’s the only discrete graphics architecture compatible with the x86 programming model, paving the way for future software and hardware integration.
Scalable Ultra Parallel Processing Architecture
Enables revolutionary floating-point performance across a wide range of computing applications:
1.48 TFLOPS of peak double-precision floating-point performance
5.91 TFLOPS of peak single-precision floating-point performance
Error Correcting Code (ECC) Memory
Helps ensure the accuracy of your computations by correcting any single or double bit error as a result of naturally occurring background radiation.
AMD PowerTune Technology
AMD PowerTune is an intelligent system that performs real-time analysis of applications that utilize a GPU3. In the event that an application is not making the most of the power available to the GPU, AMD PowerTune can improve that application's performance by raising the GPU's clock speed by up to 30% - automatically.
GPU Virtualization
Support for leading virtualization technologies, including Microsoft® RemoteFX for powering multiple users running common office applications from one GPU, as well as direct GPU pass-through support for VMware® ESXi/ESX and Citrix® XenServer that enables users running more advanced workstation graphics applications (one GPU per user model).
GeometryBoost Technology
Allows the GPU to process geometry data at a rate of twice per clock cycle resulting is a doubling in the rate of primitive and vertex processing. Triangle rates increase two-fold relative to a GPU that does not possess GeometryBoost technology.
Partially Resident Textures (PRT)
PRT can utilize absolutely enormous texture files, up to 32 terabytes large, with minimal performance impact. PRT accomplishes this by streaming small bits of these massive textures into the GPU as needed, giving compatible applications a virtually endless supply of unique texture data.
Support for OpenCL™ 1.2, DirectX® 11.1 and OpenGL 4.2
Support for latest industry standard APIs enables optimal performance in leading applications.
AMD Eyefinity Technology
Industry-leading multi-display technology enabling highly immersive and unrivaled multi-tasking experience across up to five displays4, powered by a single AMD FirePro™ S10000 graphics card.
Memory
Size/Type: 6GB GDDR5
Interface: 384-bit
Bandwidth: 480 GB/s (aggregate of dual GPUs)
Compute Performance
5.91 TFLOPS single and 1.48 TFLOPS double precision peak floating point performance
Error Correcting Code (ECC) memory support
Display Outputs
DisplayPort: Four Mini DisplayPort
DVI: One standard
Max DisplayPort 1.2 Resolution: 4096x2160
Max DisplayPort 1.1 Resolution: 2560x1600
API/Features/OS Support
DirectX® 11.1
OpenGL 4.2
OpenCL™ 1.2
Shader Model: 5.0
AMD Eyefinity Technology Support: Yes
AMD CrossFire™ Pro Support: No
OS Support: Microsoft® Windows® 8 , Windows® 7, Windows Vista®, Windows Server® 2008 R2 SP1 and Linux® (32- or 64-bit)5
Cooling/Power/Form Factor
Max Power: 375W
Slots: Two
Form Factor: full height/ full length
Bus Interface: PCIe® x16, 3.0 for optimal performance
Cooling: Active fan
System Requirements
Available PCI® Express x16 (dual slot), 3.0 for optimal performance
750W or greater power supply with two (2) 150W PCIe® AUX 8-pin power connectors
2GB system memory
Microsoft® Windows® 8 , Windows® 7, Windows Vista®, Windows Server® 2008 R2 SP1 and Linux® (32- or 64-bit)5
Internet connection for driver installation
Retail Package Contents
AMD FirePro™ S10000 server graphics card
