GE Intelligent Platforms today extended its family of rugged GPGPU (general purpose computing on a graphics processing unit) solutions with the announcement of the 6U OpenVPX NPN240 multi-processor. The NPN240 features two NVIDIA® CUDA-capable GT240 96-core GPUs, enabling it to deliver up to 750 GFLOPS peak per card slot (depending on the application). Multiple NPN240s can be linked to single or multiple hosts to create multi-node CUDA GPU clusters capable of thousands of GFLOPS.
Providing a highly computationally-dense platform, the NPN240 is ideal for demanding applications that are constrained in terms of size, weight and power. Many military/aerospace companies and organizations are finding CUDA technology exciting, as its innovative technology shows significant opportunity for a sustainable competitive advantage: it has been demonstrated to be capable of delivering significant improvements, with one company experienced a 15x increase in throughput in a radar application. Beyond this, the relative ease of CUDA application development, together with the rapidly-growing infrastructure of software development environments and CUDA-knowledgeable programmers, promises to deliver considerable gains in productivity.
The rugged NPN240 is designed to deliver game-changing levels of performance in deployed applications that lend themselves to parallel processing such as radar, sonar, and signal and sensor processing as well as state-of-the-art image, video and graphics . Typical applications include stream processing of high resolution images for deployed airborne surveillance. It is designed to be used in conjunction with the GE Intelligent Platforms IPN250 (announced March 11) or SBC620, or any other OpenVPX-compatible host single board computer. The IPN250 combines a single GT240 96-core CUDA GPU with an Intel® Core™2 Duo host processor operating at 2.26GHz and 8 GBytes of DDR3 SDRAM to deliver up to 390 GFLOPS of performance per card slot.
“As well as providing greatly increased computing density for today’s rugged deployed mission computers, the application footprint for GPGPU is growing rapidly as an increasing number of developers find innovative ways to exploit the parallel processing capabilities of the latest multicore GPUs,” said Peter Cavill, General Manager, Military & Aerospace at GE Intelligent Platforms. “GPGPU technology provides a more flexible and highly cost-effective alternative to the FPGA-centric solutions that have traditionally been applied to some of the industry’s toughest computing problems.”
The OpenVPX- and VITA48/REDI-compliant NPN240 is interoperable with other OpenVPX-compatible COTS boards to ease integration and reduce development time. Each CUDA-capable GPU node includes local DDR3 SDRAM as well as a 16-lane PCI Express® gen2 interface to the system backplane, providing maximum data throughput direct to GPU memory. Air-, spray- and conduction cooling build options are possible to allow deployment in the harshest military and aerospace environments.
The NPN240 will be available with GE Intelligent Platforms’ AXISLib high performance VSIPL DSP and math libraries for maximum code portability, enabling programmers to target their algorithm at the CPU or the GPU to optimize system loading and performance. In addition to NVIDIA’s CUDA environment, supported software includes Linux®, Windows®, Microsoft DirectX, OpenCL, OpenGL and MATLAB®, together with NVIDIA’s PureVideo® Technology and PhysX™.
About GE Intelligent Platforms
GE Intelligent Platforms is an experienced high-performance technology company and a global provider of software, hardware, services, and expertise in automation and embedded computing. We offer a unique foundation of agile and reliable technology providing customers a sustainable advantage in the industries they serve, including energy, water, consumer packaged goods, government & defense, and telecommunications. GE Intelligent Platforms is headquartered in Charlottesville, VA. For more information, visit ge-ip.com/.
PCI Express is a registered trademark of PCI-SIG. Intel is a registered trademark, and Core is a trademark, of Intel Corporation. NVIDIA and PureVideo are registered trademarks, and PhysX is a trademark, of NVIDIA Corporation. All other trademarks are the property of their respective owners.