Prev
Next
Rugged VPX-Quadro RTX
• Powered by NVIDIA Quadro RTX 3000
• Maximum 1920 CUDA cores, 30 RT cores and 240 Tensor cores
• 6GB GDDR6 192-bit with 336 GB/s
• Support up to 4 DisplayPort 1.4 displays
• NVIDIA GPUDirect, CUDA Compute version 7.5, Vulkan 1.1 support
• 5-year lifecycle availability
Overview
Aetina Quadro VPX module is the rugged CUDA-enable GPU computing module in 3U VPX. It’s compliant to OpenVPX 65, which stands for harsh environment. Powered by Turing™ -based NVIDIA Quadro, it embraces advanced GPU parallel computing and breakthrough in AI inference capability for VPX systems.
Equipped with Aetina RobustCool™ thermal system, it’s the fundamental pillar of supporting wide operating temperature. Aetina released the new generation of Quadro RTX VPX, helping developers maximize utilization to meet their GPU acceleration needs in defense, aerospace, aviation, UAV, ground vehicles and marine.
Features
● Best Leap with NVIDIA Quadro Turing GPU
The Turing GPU includes the Tensor Cores to accelerate diverse neural networks and AI with great performance. The RTX 3000 is powered by NVIDIA Quadro Turing TU106. It contains 1920 CUDA cores, 240 Tensor cores and 30 RT cores, driving the ultimate 5.3 TFLOPS peak FP32 performance and dedicated AI inference, providing more than 2X faster than the previous generation. It’s ultimate extension GPGPU processing with revolutionary rendering technology that maximizes complex visualization and signals verisimilitude of image analysis.
● Revolutionary GDDR6 Memory
With the latest GDDR6 memory and support of GPUDirect and ECC memory, NVIDIA Quadro RTX donates greater speed and numerous data communication bandwidth, smoothing I/O data access bottleneck and stabilize the data acquisition.
● NVNEC & NVDEC for 8K Video Encode
NVIDIA Quadro RTX supports NVENC and NVDEC, the latest video encode and decode technology of NVIDIA Video Codec and hardware acceleration engine, driving 8K/4K video encoding at 30 fps with uplift in bitrate.
● Support CUDA Compute version 7.5, OpenCL 1.2, OpenGL 4.6, DirectX 12 and Vulkan 1.1 API.
● Potent Conduction-Cooling Techniques
The Quadro VPX module was equipped with Aetina new invention, the RobustCoolTM thermal system. The conduction-cooling uses a copper-based heat frame to enable superior thermal conductivity efficiency and comprehensively speed heat transfer from GPU to the heat frame and the VPX chassis.
| Thermal Conductivity (W/m.K) | |
| Copper | 401 |
| Aluminum | 237 |
Specifications
Model Number
V3T3000-QRC
GPU Engine
GPU Architecture
NVIDIA Turing™ TU106
NVIDIA CUDA Cores
1920
Floating Point Performance
5.3 TFLOPS SP Peak
Memory
Memory Size
6GB GDDR6
Memory Clock
14 Gbps
Memory Interface Width
192-bit
Memory Bandwidth
336 GB/sec
Feature Support
Bus Type
PCI Express 3.0
DirectX
12
OpenGL
4.6
OpenCL
1.2
Operation System
Windows 10 64-bit
Linux 64-bit
Linux 64-bit
VITA Standards
VITA Standards
VITA 46.0 VPX Base Standard
VITA 46.4 PCI Express on VPX Fabric Connector
VITA 65 OpenVPX Architecture Framework for VPX
VITA 46.4 PCI Express on VPX Fabric Connector
VITA 65 OpenVPX Architecture Framework for VPX
Display Support
Max. Displays per Board
4
Max. Digital Resoultion
7680x4320
Display Interface
Four DisplayPort 1.4 outputs to VPX P2
Power
Board Power (TGP)
80 W
Thermal
Cooling
Conduction cooled
Mechanical & Environment
Form Factor
3U VPX
Dimensions (mm)
100 x 160 mm
Operation Temperature (Standard)
0 to +55°C
Operation Humidity
Relative Humidity 5 to 90%
Operation Temperature (Optional)
-40 to +71°C (at wedge lock)
Storage Temperature
-40 to +125°C, Relative Humidity 5 to 95%
Dimensions
.jpg)
Ordering Information
| Model Number | Description |
|---|---|
| V3T3000-QRC | VPX3U, NVIDIA Quadro RTX3000, 6GB GDDR6, Conduction-cooled, 0°C to +55°C (at Wedge-lock) |
| V3T3000-QRC-H | VPX3U, NVIDIA Quadro RTX3000, 6GB GDDR6, Conduction-cooled, -40°C to +71°C (at Wedge-lock) |
Downloads
| Model Name | File Name | OS | Version | Update Date | Download |
|---|---|---|---|---|---|
| V3T3000-QRC | Aetina_VPX_V3T3000-QRC_datasheet | v2.0 | 2022-02-14 | View |