Sapphire 31004-44-20A Fascicule
AMD FirePro
™
S10000
High-Density, High-Performance
Server Graphics
Key features:
• Two powerful GPUs
• Dual-slot form factor
• Passive thermal design
• 6GB or 12GB GDDR5 memory
• Up to 480GB/s memory bandwidth
• ECC memory support
• 5.91 TFLOPS peak single precision floating point
• 1.48 TFLOPS peak double precision floating point
• 375W maximum power
• Support for SMBus temperature reporting at boot up
• Compatible with VMware
• Dual-slot form factor
• Passive thermal design
• 6GB or 12GB GDDR5 memory
• Up to 480GB/s memory bandwidth
• ECC memory support
• 5.91 TFLOPS peak single precision floating point
• 1.48 TFLOPS peak double precision floating point
• 375W maximum power
• Support for SMBus temperature reporting at boot up
• Compatible with VMware
®
vSGA and vDGA technologies
• Support for Microsoft
®
RemoteFX shared graphics
• Compatible with Citrix
®
XenApp
™
• PCIe
®
3.0 compliant
• Support for industry standard APIs
• Designed, built and thoroughly tested by AMD
• Planned minimum two-year life cycle
• Three-year limited warranty
• Designed, built and thoroughly tested by AMD
• Planned minimum two-year life cycle
• Three-year limited warranty
Data center and IT managers are faced with many challenges: doing more with
less resources, configuring computing solutions to meet a variety of end- user
needs, flat power budgets, reigning in time spent on system support and
maintenance, preventing data leakage, and supporting multiple operating
systems and application versions. AMD FirePro
less resources, configuring computing solutions to meet a variety of end- user
needs, flat power budgets, reigning in time spent on system support and
maintenance, preventing data leakage, and supporting multiple operating
systems and application versions. AMD FirePro
™
technology is designed to help
managers meet these challenges and more. AMD FirePro
™
S10000 server graphics
deployed in the data center can help to reduce operating costs and time spent
on servicing individual systems, increase asset utilization density and maximize
processing power within the same power budget.
on servicing individual systems, increase asset utilization density and maximize
processing power within the same power budget.
High-density supercomputing
AMD FirePro S10000 server graphics are designed to meet the most demanding
performance and reliability requirements. For deployments requiring intense
processing and accuracy, like Computational Fluid Dynamics, Structural
Mechanics, Numerical Analytics and Genetic Sequencing, AMD FirePro S10000 is
a GPU compute powerhouse, delivering record-breaking peak single and double
precision floating point performance in a high-density dual-GPU form factor,
performance and reliability requirements. For deployments requiring intense
processing and accuracy, like Computational Fluid Dynamics, Structural
Mechanics, Numerical Analytics and Genetic Sequencing, AMD FirePro S10000 is
a GPU compute powerhouse, delivering record-breaking peak single and double
precision floating point performance in a high-density dual-GPU form factor,
1
and
error code correcting (ECC) memory support.
Generating an impressive 1.48 TFLOPS of peak double-precision floating-
point performance - up to 7.8 times the competing dual-GPU solution
point performance - up to 7.8 times the competing dual-GPU solution
2
-
AMD FirePro S10000 is capable of tackling the most demanding compute-
intensive, data-parallel tasks. With 6GB or 12GB of GDDR5 memory, up to
480GB/s of memory bandwidth and up to 5.91 TFLOPS of peak single precision
floating point performance, compute-intensive workloads are no match for
the AMD FirePro S10000.
intensive, data-parallel tasks. With 6GB or 12GB of GDDR5 memory, up to
480GB/s of memory bandwidth and up to 5.91 TFLOPS of peak single precision
floating point performance, compute-intensive workloads are no match for
the AMD FirePro S10000.
Additionally, AMD FirePro S10000 and the current family of AMD FirePro
™
server
graphic products support key industry standard APIs, including OpenCL
™
and
DirectX
®
for writing programs that execute across heterogeneous platforms
consisting of CPUs, GPUs and other processors.
Tackling graphics and compute-heavy work flows
As traditional workstation graphics applications evolved, they required more
graphics processing performance to help professionals create and render more
complex 3D models, designs and animations. Today, applications used by
graphics processing performance to help professionals create and render more
complex 3D models, designs and animations. Today, applications used by