ASHBURN, Va., 2 May 2016. The Curtiss-Wright Corp. Defense Solutions division in Ashburn, Va., is introducing the rugged CHAMP-GP3 general purpose graphics processor unit (GPGPU) embedded computing module for deployed radar, signals intelligence, and electro-optical signal processing.
The application accelerator 6U OpenVPX module offers the floating point processing power of the NVIDIA Maxwell architecture class GPGPUs. The board features dual 1536-core Tesla M6 MXMs high performance computing devices to provide 5 trillion floating point operations per second (teraFLOPS) of compute performance in one slot.
Designed for compute-intensive applications that require extreme processing, the CHAMP-GP3 provides a size, weight and power (SWaP) optimized solution for system designers who seek the unmatched super computing capability of a Tesla GPGPU to address applications that can't be satisfied by aggregating multiple microprocessors.
The CHAMP-GP3 is designed for use in intelligence, surveillance, and reconnaissance (ISR) and electronic warfare (EW) applications that require teraFLOPS of accelerated processing.
Related: Curtiss-Wright demonstrates air cooling for rugged embedded computing in 3D printed chassis
It supports the CUDA and OpenCL programming languages to enable designers using the CHAMP-GP3 to develop applications quickly. The module addresses the massive amounts of data generated by modern radar, signals intelligence and infrared sensors, and provides high-performance embedded computing (HPEC) in cross-cueing applications.
For more information contact Curtiss-Wright Defense Solutions online at www.curtisswrightds.com.
