The updated SOSA aligned U-C5300 multi-head GPGPU is the most powerful AI (Artificial Intelligence) enabled 3U VPX GPGPU board, providing remarkable performance in a compact and rugged form factor.

Available with powerful NVIDIA GPU options based on latest Turing architecture, the U-C5300 is ideally suited for AI Delivery, Video Analytics, Image Processing, and many other applications.

The RTX3000 includes 1920 CUDA Cores for parallel processing, 240 Tensor Cores for AI inference and 30 RT Ray-Tracing Cores for real time rendering.

The parallel processing capabilities of today’s multi-core GPUs make them ideal for both non-graphics and graphics applications with intensive computation requirements. Aitech’s U-C5300 General Purpose AI GPU (GPGPU) board provides these capabilities, as well as high-performance graphics rendering capabilities and multiple video output channels, in a rugged 3U VPX form factor.

In addition to the increased throughput offered by parallel processing, GPGPU computing also allows the CPU and OS to remain responsive even when the system is under a heavy load, by offloading the intensive operations to the GPU. GPGPU application development can be performed on a standard PC that is equipped with a GPU of the same architecture.

The U-C5300 hosts an MXM GPU module (standardized GPU form factor), and new configurations of the U-C5300 are released as higher-performance MXMs become available. The U-C5300 currently supports the NVIDIA® Quadro® RTX3000 and NVIDIA® Quadro® T1000 GPUs.

The U-C5300 operates as a peripheral board with a compatible x86 VPX host SBC, such as Aitech’s C875 (8th Gen. Intel® Xeon® E), and C877/U-C8770 (Intel® Xeon® D) 3U VPX SBCs.
The U-C5300 and the host SBC interconnect over the VPX backplane, via a high speed PCIe Gen3 link of up to 16 lanes.