Description
Overview
Model Number |
S9300 X2 |
Manufacturer |
AMD |
Product Type |
FirePro |
Memory |
8GB HBM (High Bandwidth Memory) |
Interface |
PCIe Gen 3 x16 |
Memory Bandwidth |
1 TB/s (512GB/s per GPU) |
Form Factor |
Dual Slot, Full Length, Full Height |
OS Support |
Linux 64-bit |
Warranty |
30 Days DOA Only |
Key Features
World's First GPU Accelerator with HBM MemoryThe AMD FirePro S9300 x2 Server GPU is the industry's first server GPU with 8GB of ultra-fast HBM memory and features 1 TB/s of memory bandwidth (512GB/s per GPU). HBM is a new type of memory design with low power consumption and ultra-wide communication lanes. It uses vertically stacked memory chips interconnected by microscopic wires called "through-silicon vias," or TSVs, placed directly onto the interposer, shortening the distance information has to travel between memory and processor.
Open Developer Ecosystem
AMD FirePro S9300 x2 Server GPU is fully compatible with AMD GPUOpen software stack, which includes a new open source Linux driver, built from the ground up and optimized for compute. This driver includes native support for GPU-to-GPU communication both within and between nodes.
Developers now have the option of using OpenCL or C++ to accelerate code on the GPU. The free, open source Heterogeneous Compute Compiler (HCC) leverages popular CLANG/LLVM technology and supports compilation of C++ code to x86 or GPU targets. Since a majority of developers already use C++ programming, there's no need to learn a new language to start accelerating on the GPU. The open source HCC compiler will allow developers to have single source code development for the new AMD FirePro S9300 x2 GPU.
13.9 TFLOPS of Peak Single Precision
Helps speed up time required to complete single precision floating point operations used within Simulations, Video Enhancement, Signal Processing, Video Transcoding and Digital Rendering applications where high performance takes precedence over accuracy. With the AMD FirePro S9300 x2 delivering 13.9 TFLOPS of peak single precision compute performance, one can configure a 2P server with 8GPUs to achieve over 111 TFLOPS of peak single precision compute performance.
870 GFLOPS of Peak Double Precision
Helps speed up time required to complete double precision floating point operations used within Computation Fluid Dynamics, Structural Mechanics, Reservoir Simulation and Aerodynamics applications, where numerical precision is mission critical.
8GB HBM Memory
HBM is a new type of memory design with low power consumption and ultra-wide communication lanes. It uses vertically stacked memory chips that have a total bandwidth of 1 Terabyte per second.
Payment & Security
Your payment information is processed securely. We do not store credit card details nor have access to your credit card information.