GPGPU (general purpose computing on a graphics processing unit) applications include ISR, radar and sonar.
Featuring a 384-core NVIDIA Kepler architecture-based EXK107 GPU, the new GE graphics board comes with 2GBytes of GDDR5 memory which, together with the latest NVIDIA GPU makes the GRA112 significantly more powerful than its predecessor, the GRA111. A simple, cost-effective technology insertion upgrade is available for the GRA111 model.
Availability of the pin-compatible GRA112 also provides a technology insertion upgrade path for users of the MAGIC1 rugged display computers, which currently feature the GRA111 processor.
Described as a ‘disruptive’ technology that has enabled the deployment of sophisticated military/aerospace applications significantly higher in performance but with short development periods, the GPGPU is already widely deployed in a broad range of military/aerospace applications.
The truly rugged GRA112 graphics board can be deployed confidently in some of the most challenging military/aerospace environments subject to extremes of shock, vibration, temperature and contaminants. The GRA112 is also an ideal solution for SWaP-constrained applications such as unmanned vehicles.
The NVIDIA EXK107 GPU features the company’s CUDA parallel computing architecture and delivers significant advances in performance per watt and ease of use over previous generations. The 384-core GPU is capable of 622 GFLOPS, 64 GBytes/second internal memory bandwidth and 16 GBytes/second external PCI Express bandwidth.
NVIDIA CUDA technology enables development of highly advanced high performance embedded computing solutions for military applications. Combined with Kepler GPU architecture, the technology will enable platforms such as GE’s GRA112 to deliver high levels of capability.
GE’s GPU Direct Peer-to-Peer technology allows data to stream directly onto the GPU, resulting in a 25x or more decrease in transfer latency from sensors to GPUs, which substantially increases the efficiency of GPGPU applications while simultaneously freeing up CPU resources and memory, allowing GPUs to be applied to time-sensitive applications such as electronic warfare that were previously not viable.
Key features of GRA112 graphics boards:
- Latest PCI Express Gen 3 interconnect
- Double-precision floating point hardware minimises the impact of artefacts introduced by single-precision processing for radar processing
- Higher performance enables superior throughput in applications such as video stitching and sensor fusion in wide area surveillance applications
- Allows data from multiple sensor types to build a complete picture of a large geographical area, delivering increased situational awareness and context
- Rich set of I/O helps serve most common video applications
- Dual, independent channels allow the board to drive RGB analogue component video, digital DVI 1.0, and HDMI standards
- Video input capability allows integration of sensor data using RS170, NTSC or PAL video formats
- Development environment includes support for OpenCL, OpenGL, NVIDIA PureVideo Technology (PUHD), NVIDIA PhysX and Microsoft DirectX