- Overview
- Specifications
- Downloads
- Featured Products
- Request a Quote
- Product Support
The rugged 3U OpenVPX™ form factor VPX3-493 supports NVIDIA’s size, weight, power and cost (SWAP-C) optimized 2.6 TFLOPS T1000 GPU devices.
The ascendance of GPUs in the field of high-performance computing is demonstrated in the NVIDIA Turing architecture, with many features improving GPU computing over prior generations. These features include a new unified shared memory architecture, the addition of Turing Tensor cores, support for GDDR6 memory, and many other enhancements designed to improve CUDA-based application performance and programmer productivity.
The VPX3-493 is designed to work with Intel® processor-based single board computers (SBC) such as the Curtiss-Wright VPX3-1260 8th Gen Xeon board. The x16 Gen3 PCIe interface presented by the VPX3-493 can be used with x4 and x8 capable hosts. In addition, the PCIe switch can be configured to support one x8 or two x4 downstream PCIe ports (contact the factory for more information).
The VPX3-493 also supports IPMI with on-board voltage, current and thermal sensors.
Key Features
- ISR and EW applications where TFLOPS of accelerated processing is required
- Massive data ingest of modern Radar SIGINT, EO/IR sensors
- Unparalleled HPEC performance
Applications
- ISR and EW applications where TFLOPS of accelerated processing is required
- Massive data ingest of modern Radar SIGINT, EO/IR sensors
- Unparalleled HPEC performance
Data Sheet
White Paper
Brochure
