CWCEC announces delivery of CHAMP-AV8 and VPX6-490 boards to customer. Intel 2nd Gen Core i7 SBC and NVIDIA GPGPU pair establishes new high watermark for computing density and off-card fabric bandwidth
ASHBURN, VA – December 2, 2011 — Curtiss-Wright Controls Embedded Computing, a business group of Curtiss-Wright Controls and a leading designer and manufacturer of commercial off-the-shelf (COTS) products for the rugged deployed aerospace and defense C4ISR market, has announced the first deliveries of its new CHAMP-AV8 and VPX6-490 signal and image processor boards, the industry’s highest density and highest performance fabric bandwidth VPX modules. The two modules, both designed to meet the most demanding SIGINT, COMINT and radar processing applications, together with CWCEC’s FPGA engines and FMC I/O modules, deliver the industry’s most comprehensive and highest performance single board computer (SBC) and GPGPU solutions for the most demanding HPEC applications.
The CHAMP-AV8, CWCEC’s first rugged, high performance OpenVPX™ DSP (digital signal processing) engine based on the new quad-core Intel® Core™ i7-2715QE processor, accelerates SIGINT algorithms with its 256-bit AVX floating point instruction set. The VPX6-490 GPGPU compute engine features dual NVIDIA® GPUs based on the NVIDIA Fermi architecture, and the massive compute capability of a pair of 240-core GPU devices. Together these two 6U OpenVPX processor engines comprise the heart of very high performance rugged deployable signal processor systems for aerospace and defense applications.
“We are very excited to announce that we are currently shipping our CHAMP-AV8 and VPX6-490 VPX DSP engines,” said Lynn Bamford, vice president and general manager of Curtiss-Wright Controls Embedded Computing. “We are proud to lead the embedded market with the highest performance COTS processors available today for demanding radar and signal intelligence systems.”
About the CHAMP-AV8
The CHAMP-AV8 multi-processing board brings the floating-point performance of the quad-core Intel® Core i7-2715QE processor to the OpenVPX form factor standard. The CHAMP-AV8’s dual processors deliver performance rated at up to 269 GFLOPs. With a 21GB/s (peak) DDR3 memory subsystem connected directly to the processor, the Intel® Core i7-2715QE is able to maximize the throughput of its Intel AVX vector processing units and process larger vectors at peak rates significantly greater than was possible with previous AltiVec™-based systems.
The CHAMP-AV8 is CWCEC’s first DSP engine to offer IDT’s Tsi721 Gen2 PCIe-to- Serial RapidIO® (sRIO) protocol conversion bridge, bringing the performance and bandwidth advantages of sRIO switch fabrics to the 2nd generation Intel® Core™ i7 processor-based embedded computing applications. Available in a full range of air- and conduction-cooled configurations, the rugged CHAMP-AV8 combines the significantly improved floating-point performance of the latest Intel® Core™ i7 processors, with the substantial bandwidth and system-enabling features of the 6U OpenVPX™ form factor. The board’s pair of quad-core processors feature new Intel AVX floating-point instructions, and delivers unmatched performance, rated at up to 269 GFLOPS.
Supporting the DSP engine’s floating-point performance is a 21GB/s (peak) DDR3 memory sub-system that provides ample bandwidth to simultaneously serve CPU access and streaming I/O from its sRIO and PCIe interfaces.
The CHAMP-AV8 incorporates the enhancements of the OpenVPX (VITA 65) standard with a complete suite of data plane, expansion plane, and control plane interfaces. Supporting Gen2 SRIO and Gen2 PCIe interfaces, the CHAMP-AV8 offers triple the bandwidth of first generation VPX products with up to 32 GB/s of fabric performance, thus ensuring that application performance can scale commensurately with the much higher CPU performance.
CHAMP-AV8 Key Features:
· High-performance floating-point processing for digital signal processing (DSP) and general purpose applications
· 6U OpenVPX™ compliant
· Two quad-core Intel® Core™ i7 processors
· Intel® AVX 256-bit floating-point vector unit
· 269 GFLOPs peak computing performance
· Up to 8 Gbytes DDR3 SDRAM with ECC per processor\
· Gen2 Serial RapidIO® (sRIO) data plane interface
· Gen2 PCI Express® (PCIe) expansion plane interface
· DMA offload engines
· One XMC mezzanine site
· Power management features
· VxWorks® BSP and driver suite
· Linux BSP and driver suite
· Rugged air-cooled and conduction-cooled versions
Extensive Software Support
The CHAMP-AV8 is supported with an extensive suite of software including support for Wind River VxWorks and Linux operating environments. Additional software support includes communication middleware supporting MPI and OFED for Linux, Curtiss Wrights Inter-processor communications (IPC) for VxWorks and Curtiss-Wright Controls’ Continuum Vector AVX-optimized signal processing library for Linux and VxWorks.
About the VPX6-490
The VPX6-490 GPU Application Accelerator is CWCEC’s first general purpose graphics processing unit (GPGPU)- multi-core engine. The VPX6-490 features dual NVIDIA® GPUs based on the NVIDIA Fermi architecture, each with 240 CUDA™ cores. Integrated into a High Performance Embedded Computing (HPEC) subsystem, VPX6-490 functions as a co-processor attached to a host Intel®-processor board and takes advantage of the new PCIe Expansion Plane definitions in the VITA 65 OpenVPX standard to provide off the shelf backplane support for high-speed interconnection between pairs of SBC/GPU. The combination of 2nd Generation Intel processors, gen2 PCI Express interconnect and 480 NVIDIA CUDA cores raises the performance bar for embedded High Performance Embedded Computing (HPEC) systems for demanding military digital signal processing (DSP) applications such as C4ISR, EO/IR, and SatCom.
The VPX6-490 uniquely brings NVIDIA’s new 240 core graphics processors to the rugged deployed environment, providing system integrators with the highest performance GPGPU embedded building blocks available. The VPX6-490 takes full advantage of GPUs based on the NVIDIA Fermi architecture. Designed for high performance computing, the newest generation of NVIDIA processors feature larger internal shared memories, a completely new L2 cache, unified memory addressing and many other enhancements to improve CUDA-based applications performance and improve programmer productivity. The VPX6-490 supports its dual high performance GPU processors with a 2 Gbyte, 256-bit wide, 80 Gbyte/s GDDR5 memory subsystem designed to eliminate data bottlenecks and support large signal processing datasets into the onboard memory. Each GPU device supports a full 16-lane Gen2 PCIe interface to the backplane, supporting the maximum possible bandwidth between host and GPU. The VPX6-490 also supports 8-lane and 4-lane PCIe interfaces.
VPX6-490 Performance Features:
· 6U OpenVPX™ GPGPU Application Accelerator
· 80 Gbyte/s memory bandwidth
· Enhanced GPGPU performance based on NVIDIA Fermi architecture:
– 240 CUDA processor cores
– 2 Gbytes GDDR5 memory
– 256-bit memory interface
– 16-lane Gen2 PCIe interface
– Dual DVI graphics outputs
· CUDA 4.0 library support
· Rugged air-cooled and conduction-cooled variants
· Temperature sensors
· IPMI
· 12V power
Curtiss-Wright Continuum HPEC Subsystem Components
The CHAMP-AV8 and VPX6-490 are key elements of CWCEC’s Continuum High Performance Embedded Computing (HPEC) initiative. Continuum HPEC systems consist of a large number of distributed processors, IO, and software stacks connected by a low latency system fabric. HPEC capabilities are developed in our Ashburn, VA HPEC Center of Excellence. With scalable architectures, dataflow modeling and configuration validation, Curtiss-Wright’s Continuum HPEC customers can source embedded supercomputing platforms that integrate Intel®-based multi-processor boards with AVX, GPGPU co-processors, Xilinx® Virtex® 6 FPGAs, SRIO and Ethernet switching with Open Standard software solutions including VxWorks®, and Linux with OpenMPI and OFED software interfaces. Supported products include Curtiss-Wright’s CHAMP-FX3 Virtex6 FPGA board, the CHAMP-AV8 dual 2nd Generation Core® i7-based multiprocessor board, the VPX6-1956 2nd Generation Core i7 SBC, the VPX6-6902 sRIO/Ethernet switch, and the VPX6-490 GPGPU module. OpenVPX™ enclosures are supported including a small 5-6 slot and 19” rack 16-slot air-cooled Chassis.
For price and availability, please contact the factory. Click here, to learn more about our family of digital signal processing engines.
For editorial information regarding Curtiss-Wright Controls Embedded Computing products or services, contact John Wranovics, public relations director, Curtiss-Wright Controls Embedded Computing, Tel: (925) 640-6402; email. [email protected].
Inquiries: Please forward all Sales and reader service inquiries to Jerri-Lynne Charbonneau, Curtiss-Wright Controls Embedded Computing, Tel: (613) 254-5112; Fax: (613) 599-7777; e-mail: [email protected].
About Curtiss-Wright Controls Embedded Computing
Curtiss-Wright Controls Embedded Computing is the industry’s most comprehensive and experienced single source for embedded solutions, ranging from Processing, Subsystems, Data Communication, DSP, and Video & Graphics to the most advanced board level components and fully integrated custom systems. The Embedded Computing group serves the defense, aerospace, commercial and industrial markets and is part of Curtiss-Wright Controls Inc.
About Curtiss-Wright Controls, Inc.
Headquartered in Charlotte, North Carolina, Curtiss-Wright Controls is the motion control segment of Curtiss-Wright Corporation. With manufacturing facilities around the world, Curtiss-Wright Controls is a leading technology-based organization providing niche motion control products, subsystems and services internationally for the aerospace and defense markets. For more information, visit www.cwcontrols.com/.