3U OpenVPX™ VPX3-4936 GPU module delivers 17.7 TFLOPS and 68 dense/136 sparse Tensor TOPS, expanding Curtiss-Wright’s family of high-performance AI/ML solutions
AUSA 2022, Walter E. Washington Convention Center, Washington D.C. (Hall A/B/C - Booth 1309) – October 10, 2022 – Curtiss-Wright’s Defense Solutions division, a leading supplier of rugged ISR and EW processing modules and systems, today introduced the VPX3-4936, its first 3U OpenVPX GPGPU processor module to feature the combination of NVIDIA’s Ampere GPU and a configurable Gen4 PCIe Switch. The NVIDIA Ampere architecture offers a significant boost in performance and efficiency over the previous NVIDIA Turing™ generation, including more flexible concurrent execution of floating point and integer streams. The Ampere device’s third generation Tensor cores deliver up to 4x acceleration of AI/ML algorithms and its next generation RT cores and CUDA core architecture provide 2x performance compared to the previous generation. While delivering close to 18 TFLOPS FP32 peak performance and 68 dense/136 sparse Tensor TOPS, the NVIDIA Ampere also significantly improves power efficiency, yielding an extremely efficient 154 GFLOPS per Watt. The module’s PCIe Gen4 architecture also doubles the host interface bandwidth, eliminating data throughput bottlenecks. Pin-compatible with Curtiss-Wright’s Turing architecture VPX3-4935, the VPX3-4936 enables system designers to boost math-intensive processing algorithms without increasing size, weight, and power (SWaP). The board’s PCIe architecture also supports Non-Transparent Bridging (NTB) and daisy chain options for system flexibility.
The rugged VPX-4936 module is designed in compliance with the U.S. Army’s C5ISR/EW Modular Open Suite of Standards (CMOSS) and is aligned with the SOSA Technical Standard to support compute-intensive ISR and EW systems. This module is ideal for use in applications that require significant computational performance, such as the accelerated processing of tensor/matrix computations needed for deep learning neural network inferencing used in deployed artificial intelligence (AI) and machine learning (ML) systems. Example applications include ISR, EW, High-performance RADAR, SIGINT, EO/IR, sensor fusion, and autonomous platforms
“The VPX3-4936 enables system developers to leverage more complex AI and signal processing algorithms and applications to support compute intensive ISR applications on deployed MOSA systems,” said Chris Wiltsey, Senior Vice President and General Manager, Curtiss-Wright Defense Solutions. “We continue to expand our SOSA aligned product portfolio to deliver the most comprehensive range of open standards based system elements for EW and ISR applications, with the highest performance GPU, DSP, and FPGA modular solutions.”
Complete System-Level HPEC Solutions
The VPX3-4936 module is fully interoperable with Curtiss-Wright’s broad family of SOSA aligned and non-SOSA aligned 3U OpenVPX solutions. For example, SWaP-constrained systems can pair a single VPX3-4936 with the Intel® Xeon® D processor-based CHAMP-XD3 DSP engine, which together can augment the powerful sensor processing capabilities of Curtiss-Wright’s Xilinx® FPGA-based transceiver modules.
To download the VPX3-4936 product sheet click here.
For additional information about Curtiss-Wright Defense Solutions products, please visit www.curtisswrightds.com, LinkedIn, and Twitter @CurtissWrightDS.
About Curtiss-Wright Corporation
Curtiss-Wright Corporation (NYSE:CW) is a global integrated business that provides highly engineered products, solutions and services mainly to Aerospace & Defense markets, as well as critical technologies in demanding Commercial Power, Process and Industrial markets. We leverage a workforce of 8,000 highly skilled employees who develop, design and build what we believe are the best engineered solutions to the markets we serve. Building on the heritage of Glenn Curtiss and the Wright brothers, Curtiss-Wright, headquartered in Davidson, North Carolina, has a long tradition of providing innovative solutions through trusted customer relationships. For more information, visit www.curtisswright.com.
###
Note: Trademarks are property of their respective owners.
This press release contains forward-looking statements made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. Such statements, including statements relating to Curtiss-Wright’s expectations of a continued relationship with an existing customer, the continued success of this embedded security technology, the performance of its products and the future opportunities associated with this technology, are not considered historical facts and are considered forward-looking statements under the federal securities laws. Such forward-looking statements are subject to certain risks and uncertainties that could cause actual results to differ materially from those expressed or implied. Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date hereof. Such risks and uncertainties include, but are not limited to: a reduction in anticipated orders; an economic downturn; changes in competitive marketplace and/or customer requirements; a change in US and Foreign government spending; an inability to perform customer contracts at anticipated cost levels; and other factors that generally affect the business of aerospace, defense contracting, marine, electronics and industrial companies. We expressly disclaim any current intention to update publicly any forward-looking statement after the distribution of this release, whether as a result of new information, future events, changes in assumptions or otherwise. Please refer to the Company's current SEC filings under the Securities Exchange Act of 1934, as amended, for further information.
John Wranovics
Director of Public Relations
John Wranovics has over thirty years of experience in managing media relations and the promotion of high technology products. He has been with Curtiss-Wright since 2003. He has a degree in English Literature from the University of California, Berkeley.