Fast processing for display graphics or math/vector acceleration engine
A GPU complements a processor to offer display graphics or can serve as a math/vector acceleration engine. As a display processor, a GPU supports real-time graphics using OpenGL and media codecs such as MPEG and H.254, and can drive multiple displays simultaneously. When used as a math/vector accelerator, a GPU can perform floating-point calculations using hundreds or thousands of parallel floating-point units via the use of OpenCL and CUDA software frameworks. In data-intensive applications, such as imaging enhancement and mosaicking, GPUs can stitch together input from multiple sensors or process radar data faster than general-purpose CPUs.
Enabling AI at the Network Edge of the Battlefield
This white paper discusses how modern AI technology can benefit the warfighter as embedded solutions at the network edge are deployed for military and aerospace platforms.
GPUs provide
- Real-time graphics using OpenGL, MPEG, and H.254 to drive multiple displays
- Fast processing of radar data
- Ability to gather and process input from multiple sensors
- Acceleration of math-intensive algorithms using floating point calculations with OpenCL and CUDA
Curtiss-Wright works closely with NVIDIA and WOLF Advanced Technology to select GPU processors for demanding aerospace and defense applications. We use industrial qualified processors to meet the high-reliability requirements that Curtiss-Wright products demand for the long-term success of your programs. Our NVIDIA GPU processing capabilities include rugged mission computing and modules are available in 3U, 6U and mezzanine form factors.