Skip to product information
1 of 1

Cisco

Cisco UCSX-GPU-L40 | NVIDIA L40 GPU, 48GB GDDR6, 300W, PCIe Gen4 x16

SKU:UCSX-GPU-L40

Stock Status: Enquire

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

Description

The Cisco UCSX-GPU-L40 is a full-height, full-length dual-slot GPU accelerator designed for Cisco UCS X-Series modular systems. Built on the NVIDIA Ada Lovelace architecture with 48GB GDDR6 ECC memory and 18,176 CUDA cores, it delivers 90.5 TFLOPS FP32 performance for AI inference, virtual workstations, 3D rendering, and graphics-intensive workloads. The 300W passive-cooled card includes PCIe Gen4 x16 connectivity and supports NVIDIA vGPU for multi-tenant virtualization in enterprise data center environments.

Features

NVIDIA Ada Lovelace architecture with 76.3 billion transistors on 4nm process
- 18,176 CUDA cores for parallel computing and AI workloads
- 4th generation Tensor Cores optimized for AI training and inference
- 3rd generation RT Cores for hardware-accelerated real-time ray tracing
- 48GB GDDR6 ECC memory with 864 GB/s bandwidth for large datasets and complex models
- PCIe Gen4 x16 interface with 64 GB/s bidirectional bandwidth
- NVIDIA vGPU support for virtual workstations and multi-tenant environments
- Hardware AV1 encode/decode with triple NVENC/NVDEC engines for video pipelines
- Support for 4x DisplayPort outputs for direct monitor connectivity
- 300W TDP with passive cooling design for rack-optimized thermal management
- Compatible with Cisco UCS Manager for centralized GPU resource management
- SR-IOV support for direct GPU passthrough to virtual machines
- Memory bandwidth of 864 GB/s through 384-bit memory interface

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 L40 based on Ada Lovelace architecture
- Memory: 48GB GDDR6 with ECC
- Memory Bandwidth: 864 GB/s
- CUDA Cores: 18,176
- Tensor Cores: 568 (4th generation)
- RT Cores: 142 (3rd generation)
- FP32 Performance: 90.5 TFLOPS
- Interface: PCIe Gen4 x16 (64 GB/s bidirectional)
- Power Consumption: 300W TDP
- Form Factor: Full-height, full-length (FHFL), dual-slot, 10.5 inch
- Cooling: Passive
- Display Outputs: 4x DisplayPort
- Platform Compatibility: Cisco UCS X410c M7 compute nodes with X440p PCIe node
- Virtualization: NVIDIA vGPU support for virtual workstations and multi-tenant environments

FAQs

Q: What Cisco platforms support the UCSX-GPU-L40?
A: The UCSX-GPU-L40 is designed for Cisco UCS X-Series modular systems, specifically the UCS X410c M7 compute nodes when configured with the X440p PCIe node. It is a full-height, full-length dual-slot card that requires proper chassis and power cable configuration.

Q: What workloads is the L40 GPU optimized for?
A: The L40 is optimized for AI inference, virtual workstation deployments, 3D rendering, real-time ray tracing, graphics-intensive applications, and video encoding/transcoding. It features hardware AV1 encode/decode capabilities and supports vGPU software for multi-user virtual desktop infrastructure.

Q: Does the UCSX-GPU-L40 include ECC memory?
A: Yes, the L40 includes 48GB of GDDR6 memory with Error-Correcting Code (ECC) support, providing enhanced reliability for production data center environments and mission-critical workloads where data integrity is essential.

Q: Can the L40 support multiple virtual machines?
A: Yes, the L40 supports NVIDIA vGPU technology, enabling GPU resources to be shared across multiple virtual machines or users. This makes it ideal for virtual workstation deployments, multi-tenant cloud environments, and centralized graphics rendering.

Q: What is the difference between the L40 and L40S?
A: The L40 has a 300W TDP and delivers 90.5 TFLOPS FP32 performance, while the L40S has a higher 350W TDP with nearly double the Tensor Core performance for AI training and generative AI workloads. Both share the same Ada Lovelace architecture and 48GB GDDR6 memory capacity.

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