Skip to product information
1 of 1

Cisco

Cisco UCSC-GPU-A40-D | A40 Graphics Card, 48GB GDDR6 ECC, 300W, PCIe 4.0

SKU:UCSC-GPU-A40-D

Stock Status: Enquire

Request Quote
Sale Sold out
Taxes included. Shipping calculated at checkout.

Description

The Cisco UCSC-GPU-A40-D is an NVIDIA A40 GPU accelerator designed for data center visual computing, AI acceleration, and compute-intensive workloads. Built on the NVIDIA Ampere architecture, it delivers 48GB of GDDR6 ECC memory, 10,752 CUDA cores, and PCIe 4.0 x16 connectivity. The passive dual-slot form factor is optimized for rack-mounted servers requiring high-performance graphics, deep learning, and virtualized desktop infrastructure deployments.

Features

NVIDIA Ampere architecture with 2nd generation RT cores and 3rd generation Tensor cores
- 48GB GDDR6 memory with ECC for mission-critical reliability
- 10,752 CUDA cores for parallel processing and compute acceleration
- PCIe 4.0 x16 interface with backward compatibility to PCIe 3.0
- Real-time ray tracing for photorealistic rendering
- Multi-instance GPU (MIG) technology for partitioning into up to 7 independent instances
- Support for NVIDIA vGPU software for virtualized graphics and compute
- NVLink connectivity to pair two A40s for 96GB combined memory
- Passive cooling design for integration into server-managed airflow environments
- Secure boot with hardware root of trust for firmware integrity
- Support for NVIDIA RTX, CUDA, cuDNN, TensorRT, and OpenCL frameworks
- Three DisplayPort 1.4 outputs (configurable, disabled by default for virtualization)
- Dual-slot form factor optimized for high-density server deployments

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

Collapsible content

Technical Specifications

FAQs

Technical Specifications

GPU: NVIDIA A40 (Ampere architecture)
- Memory: 48GB GDDR6 with ECC
- CUDA Cores: 10,752
- Tensor Cores: 336 (3rd generation)
- RT Cores: 84 (2nd generation)
- Memory Interface: 384-bit
- Memory Bandwidth: 696 GB/s
- Power Consumption: 300W TDP
- Form Factor: Passive dual-slot, full-height/full-length (FHFL)
- Interface: PCIe 4.0 x16 (backward compatible with PCIe 3.0)
- Display Outputs: 3x DisplayPort 1.4 (physical connectors disabled by default for virtualization)
- NVLink Support: Yes (pair two A40s for 96GB combined memory)
- Cooling: Passive heatsink design

FAQs

Q: What workloads is the UCSC-GPU-A40-D designed for?
A: The A40 is optimized for data center visual computing, AI and deep learning training and inference, 3D rendering, virtual workstations (VDI/DaaS), simulation, and compute-intensive HPC applications. It combines professional graphics capabilities with powerful AI acceleration.

Q: Is this GPU compatible with Cisco UCS servers?
A: Yes, the UCSC-GPU-A40-D is specifically designed for Cisco UCS C-Series rack servers. It requires PCIe x16 support and includes a unique SBIOS ID for integration with Cisco CIMC and UCSM management platforms. Verify server compatibility and available PCIe slot configuration before deployment.

Q: Does the A40 support virtualization?
A: Yes, the A40 is configured for virtualization by default and supports NVIDIA vGPU software, including RTX Virtual Workstation (vWS) and Virtual Compute Server (vCS). Physical display connectors are disabled by default but can be enabled if required for direct display output.

Q: What is the difference between UCSC-GPU-A40-D and standard A40?
A: The UCSC-GPU-A40-D is Cisco's qualified version of the NVIDIA A40 for UCS servers. It includes Cisco-specific firmware integration and is tested for compatibility with UCS hardware and management software. All Cisco GPU cards must be procured from Cisco to ensure proper system integration.

Q: Can I pair multiple A40 GPUs with NVLink?
A: Yes, you can connect two A40 GPUs using NVLink to scale memory capacity to 96GB and improve multi-GPU performance for supported applications. NVLink support depends on your specific software application, so verify compatibility with your application vendor before deployment.

Recently Viewed

  • 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