WOLF-2538
Overview
Dual NVIDIA RTX 5000 Ada HPEC with PCIe Gen4 Switch
The VPX6U-AD5000E-DUAL-VO module includes two NVIDIA RTX 5000 Ada embedded GPUs and a PCIe Gen4 switch in a rugged 6U VPX module. The NVIDIA RTX 5000 embedded GPU provides advanced processing capabilities for high performance embedded computing (HPEC) and artificial intelligence (AI) data processing.
The NVIDIA Ada architecture includes CUDA cores for HPEC, 4th generation Tensor cores for AI and data science computations, and 3rd generation Ray Tracing (RT) cores for visually accurate rendering. The Ada GPU uses a new TSMC 4N NVIDIA Custom Manufacturing Process which increases efficiency. The denser Ada GPUs have more CUDA and Tensor cores operating at higher clock frequencies at the same power, delivering significantly more performance per watt compared to the previous generation GPUs. For example, the RTX 5000 Ada GPU can achieve up to four times the GFLOPS/W compared to the RTX 5000 Turing GPU used in the WOLF-2348.
Unlocking the best performance requires the best cooling capability. WOLF’s advanced cooling technology is designed to move heat using a low weight, high efficiency path from the GPU die to the wedgelocks.,
Block Diagram
The block diagram shows a 6U VPX dual-GPU architecture with two RTX 5000 Ada GPUs connected through an onboard PCIe Gen4 switch. It highlights the high-bandwidth PCIe paths for balanced data delivery to each GPU, peer movement between endpoints through the switch fabric, and uplinks to the VPX backplane.
Features
WOLF-2538 is a rugged 6U VPX module that integrates two NVIDIA RTX 5000 Ada embedded GPUs with a configurable PCIe Gen4 switch for high-density embedded AI and HPEC acceleration with flexible VPX backplane integration. It includes 16 GB GDDR6 256-bit VRAM per GPU with ECC support and provides up to 8 DisplayPort outputs with options for HDMI by configuration. Module power is configurable from 150W to 280W, and the design supports conduction-cooled, or liquid flow-through operation with SOSA aligned profile support 10.6.4.
GPU Features
- Ada GPU parallel processing:
- CUDA Toolkit 12, CUDA Compute capability 8.9
- CUDA-X AI and CUDA-X HPEC libraries
- OpenCL 3.0, DirectX 12 Ultimate, OpenGL 4.6, OpenGL ES 3.2, Vulkan 1.2
- 304 Tensor Cores (4th Gen) per GPU
- 76 Ray Tracing cores (3rd Gen) per GPU
- NVENC (8th Gen) and NVDEC (5th Gen) with up to 8K video encoding and hardware decoding support
Connectivity / System Management
- IPMI system management
- NVIDIA GPUDirect RDMA support
- Configurable PCIe Gen4 switch
- Linux and Windows drivers
- NVIDIA driver support requires the following host CPU: Intel H/HX/P/PX/S Class, AMD H/HS Class
Mechanical / Open Systems Architecture
- High level of ruggedization:
- Rugged conduction cooled (CC), air cooled (AC), or liquid flow through (LFT)
- Operating temperature: CC: - 40°C to +85°C, optimized to 70°C, AC: - 40°C to +71°C, optimized to 60C, LFT: - 40°C to +70°C (system dependent)
- Vibration (sine wave): 10G peak, 5 - * Operational Shock: 40G peak
- Dimensions: 160mm x 233mm x 25.4mm
- Weight: AC: 1.46 kg, CC: 2.49 kg, LFT: 1.62 kg
- ANSI/VITA 48, 65 (VPX- REDI, OpenVPX)
- SOSA Aligned profile support 10.6.4
Specifications
Key configuration and performance parameters for WOLF-2538 are summarized below for quick comparison. Values shown reflect the standard module configuration; contact WOLF if you need a tailored build or a specific payload profile.
Configuration Guide
The following table defines a series of common order codes for the VPX6U-AD5000E-DUAL-VO module. The asterisks denote characters of the part number that are defined based on common configuration options.
Some common configuration options for this module include:
- Display interfaces
- Conformal coating type
- COTS, MCOTS, or locked
- PCIe switch configuration
- OpenVPX or SOSA aligned configuration
6U VPX Ada RTX5000 sample configurations
Contact Sales for the latest Ordering Numbers and available options
Representative ordering numbers
| Ordering Number | Description |
|---|---|
| 253823-F***-***VPX6vA0 | 6U VPX, air cooled, 1", dual NVIDIA RTX5000 Ada, 8x video outputs |
| 253833-F***-***VPX6vA0 | 6U VPX, conduction cooled, 1", dual NVIDIA RTX5000 Ada, 8x video outputs |
| 2538L3-F***-***VPX6vA0 | 6U VPX, liquid flow-through, 1", dual NVIDIA RTX5000 Ada, 8x video outputs |
| 253823-F***-***VPX6vA0 | 6U VPX, air cooled, 1", single NVIDIA RTX5000 Ada, 4x video outputs |
| 253833-F***-***VPX6vA0 | 6U VPX, conduction cooled, 1", single NVIDIA RTX5000 Ada, 4x video outputs |
| 2538L3-F***-***VPX6vA0 | 6U VPX, liquid flow-through, 1", single NVIDIA RTX5000 Ada, 4x video outputs |
Related Products
- NVIDIA RTX 5000 Blackwell (GB203)
- 10496 CUDA Cores, 320 Tensor Cores
- 24 GB GDDR7 VRAM
- PCIe Gen 5
- SOSA Payload
- Module power 190–320 W
- NVIDIA RTX 5000 Turing (TU104)
- 3072 CUDA Cores, 384 Tensor Cores
- 16 GB GDDR6 VRAM
- PCIe Gen 3
- SOSA Payload
- Module power 100–300 W
FAQ
Choose WOLF-2538 when you want a dual RTX 5000-class 6U module with strong compute density and multi-display capability, but your platform is standardized around PCIe Gen4 and/or you are optimizing for power, qualification continuity, or system compatibility.
WOLF-2638 is typically preferred when your platform can take advantage of PCIe Gen5 bandwidth headroom and you want the newest-generation GPU baseline.
If the backplane and host ecosystem are Gen4-limited, WOLF-2538 can be the more practical integration choice while still providing dual-GPU density.
It provides the fabric needed to balance PCIe traffic across both GPUs and align to different platform topologies.
In practice it helps with:
- Consistent feeding of both GPUs from the host without one becoming starved
- Cleaner peer/endpoint movement through a centralized PCIe fabric
- Integration flexibility when chassis lane routing differs between programs
The switch is most valuable when you expect multiple system variants or when both GPUs must be fed heavily and consistently.
Confirm these early to prevent late-stage chassis/backplane changes:
- Cooling method: supports conduction-cooled (CC) or liquid flow-through (LFT) (and air-cooled variants are also referenced in configuration examples). Ensure your chassis matches the selected cooling model.
- Slot profile alignment: SOSA aligned profile support 10.6.4—confirm your backplane/slot is designed for that profile family.
- Display routing plan: supports up to 8 DisplayPort outputs with HDMI option by configuration. Decide early how displays are routed (rear I/O expectations) so harnessing and rear transition planning are stable.
For dual-GPU modules, cooling capacity and backplane routing are typically the two dominant integration constraints.
Key ordering-time decisions are integration-driven:
- Cooling variant (CC vs LFT vs AC builds): determines chassis compatibility and sustained power capability.
- PCIe switch configuration: affects PCIe lane allocation and how traffic is balanced across two GPUs.
- Display interface configuration: number of enabled outputs (up to 8 DP) and HDMI option requirements.
- Power threshold (150W-280W envelope): must match chassis thermal headroom for sustained performance.
- COTS/MCOTS/locked and conformal coating: drives qualification approach and configuration control.
Contact Sales for the latest Ordering Numbers and available options.