• Cisco HX-GPU-A40 gold graphics card with black cooling fins
1 1

Cisco

Cisco HX-GPU-A40 | NVIDIA A40 GPU, 48GB GDDR6 ECC, 300W Passive, PCIe 4.0

HX-GPU-A40

Stock Status: Enquire

Request Quote
Sale Sold out
Shipping calculated at checkout.

 Trade supply only. Server & GPU quote requests need a company name, ABN and work email address. We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 

20-min response

Need a quote or bulk pricing? Answered within 20 minutes, Sydney business hours.

Custom Built Servers

Need a different spec, quantity or configuration?

We build configured-to-order refurbished servers from Cisco, Dell, HPE, NVIDIA Lenovo and Supermicro. Tell us what you need and we'll come back with options within 24 hours.

Configure your server now →

Description

The Cisco HX-GPU-A40 features the NVIDIA A40 GPU with 48GB of GDDR6 ECC memory, delivering professional-grade visual computing and AI acceleration for demanding data center workloads. Built on the Ampere architecture with 10,752 CUDA cores, 336 third-generation Tensor Cores, and 84 second-generation RT Cores, this passive-cooled GPU excels in virtual workstations, rendering, AI training, and simulation applications. The 300W TDP and PCIe Gen4 x16 interface provide efficient performance for server and hyperconverged infrastructure deployments.

Features

NVIDIA Ampere Architecture with 10,752 CUDA cores for parallel processing
- 48GB GDDR6 ECC memory protects against data corruption in compute workloads
- 336 third-generation Tensor Cores accelerate AI training and inference
- 84 second-generation RT Cores enable real-time ray tracing and rendering
- PCIe Gen 4.0 x16 interface with 31.5 GB/s bidirectional bandwidth
- 696 GB/s memory bandwidth supports large dataset processing
- NVLink support for dual-GPU configurations up to 96GB total memory
- Passive thermal design optimized for high-density server deployments
- NVIDIA vGPU software support for virtual workstation and compute instances
- SR-IOV capability for GPU virtualization and multi-tenant environments
- TensorFloat-32 (TF32) precision for accelerated AI workloads
- Mixed-precision Tensor Core operations (FP16, BF16, INT8, INT4)
- Concurrent execution of ray tracing, shading, and denoising operations
- 3x DisplayPort outputs for visualization applications (configurable)
- CUDA, OpenCL, DirectX 12, Vulkan, and OpenGL API support
- PCIe Gen3 backward compatibility for flexible deployment

Warranty

All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability.

All technical data should be verified on the manufacturer data sheets.

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

GPU Architecture: NVIDIA Ampere
- GPU Memory: 48GB GDDR6 with ECC
- Memory Bandwidth: 696 GB/s
- Memory Interface: 384-bit
- CUDA Cores: 10,752
- Tensor Cores: 336 (3rd generation)
- RT Cores: 84 (2nd generation)
- Peak FP32 Performance: 37.4 TFLOPS
- Peak TF32 Tensor Performance: 74.8 TFLOPS (149.6 TFLOPS with sparsity)
- System Interface: PCIe Gen 4.0 x16
- NVLink Support: 112.5 GB/s bidirectional (2-way configuration)
- Max Power Consumption: 300W
- Cooling: Passive heatsink (server-optimized)
- Form Factor: Dual-slot, full-height, full-length
- Display Outputs: 3x DisplayPort (configurable)
- Compatible Systems: Cisco UCS C-Series, HyperFlex HX-Series (M6 and later nodes)

FAQs

Q: What workloads is the NVIDIA A40 optimized for?
A: The A40 is designed for professional visual computing, AI training and inference, 3D rendering with ray tracing, virtual workstations via NVIDIA vGPU software, scientific simulation, CAD/CAE workflows, and data science applications requiring large memory capacity and ECC protection.

Q: Can I connect multiple A40 GPUs together?
A: Yes, two A40 GPUs can be connected using NVIDIA NVLink to scale memory from 48GB to 96GB and increase GPU-to-GPU bandwidth to 112.5 GB/s bidirectional. This requires application support for NVLink technology and compatible server riser configuration.

Q: Is this GPU compatible with virtualization platforms?
A: Yes, the A40 fully supports NVIDIA vGPU software including RTX Virtual Workstation (vWS), Virtual Compute Server (vCS), and vPC/vApps, enabling multiple virtual workstation instances per GPU with professional graphics and compute acceleration for remote users.

Q: What is the difference between this and consumer RTX GPUs?
A: The A40 is a data center GPU with 48GB ECC memory for data integrity, passive cooling for server environments, optimized drivers for professional applications, PCIe Gen4 support, NVLink connectivity, and certification for virtualization and 24/7 operation in enterprise workloads.

Q: Does this GPU require special installation considerations?
A: The A40 requires a PCIe Gen4 x16 slot, adequate server airflow for passive cooling, and up to 300W power delivery. For Cisco UCS and HyperFlex systems, it requires specific riser configurations and SBIOS compatibility. Check server datasheet for GPU slot support and maximum memory limitations when using GPUs.

  • Request a Quote

    Looking for competitive pricing? Submit a request, and our team will provide a tailored quote that fits your needs.

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Call us Now  

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Contact us