Skip to product information
1 of 1

Cisco

Cisco UCSX-GPU-H100-80 | UCS X-Series H100 GPU Accelerator, 80GB, 350W FHFL

SKU:UCSX-GPU-H100-80

Request Quote
Sale Sold out
Shipping calculated at checkout.

Stock Status: Enquire

 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 We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 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 UCSX-GPU-H100-80 is an NVIDIA H100 Tensor Core GPU accelerator with 80GB of memory, offered by Cisco as a GPU option for the UCS X-Series modular system. It installs in the UCS X440p PCIe Node and connects to a UCS X210c compute node through the X-Fabric modules, making it suited to AI training and inference, HPC and data analytics workloads.

This is a dual-slot, full-height full-length (FHFL) PCIe card rated at 350W with passive cooling, so it relies on chassis airflow rather than an onboard fan. It fits the UCS X440p using Riser A (UCSX-RIS-A-440P), which takes one dual-slot GPU per riser. The host node supports two x16 FHFL dual-slot cards or four x8 FHFL single-slot cards. The 80GB, 350W configuration is distinct from the UCSX-GPU-H100-NVL, which is rated at 94GB and 400W.

Deployment needs a UCS X9508 chassis with both UCS X9416 X-Fabric modules, plus UCS 6454, 64108 or 6536 Fabric Interconnects. Management is through Cisco Intersight or UCS Manager 4.3(4) or later, with Intersight Managed Mode requiring at least an Essentials licence per server.

Enquire with XS Network Tech for new or used pricing and availability.

Features

- Hopper-generation NVIDIA H100 Tensor Core GPU option for UCS X-Series modular systems
- Up to 16 GPUs per UCS X9508 chassis using UCS X440p PCIe Nodes
- Up to two additional GPUs can be added directly to each UCS X210c compute node
- PCIe connectivity delivered through UCS X-Fabric modules
- Managed through Cisco Intersight and UCS Manager
- Passive design relies on chassis airflow, with no onboard fan

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

- Cisco part number: UCSX-GPU-H100-80
- GPU: NVIDIA H100 Tensor Core GPU
- Memory: 80GB
- Power rating: 350W
- Cooling: Passive (chassis airflow)
- Form factor: Dual-slot, full-height full-length (FHFL) PCIe card
- Riser option: Riser A, UCSX-RIS-A-440P (1x dual-slot GPU per riser)
- Host node: Cisco UCS X440p PCIe Node
- Chassis: Cisco UCS X9508 Chassis
- X-Fabric modules: Cisco UCS X9416 (both required for PCIe connectivity)
- Fabric Interconnects: Cisco UCS 6454, UCS 64108 and UCS 6536
- Management: Cisco Intersight and UCS Manager (UCSM) 4.3(4) or later
- Operating temperature: 10 to 35 C (50 to 95 F)
- Non-operating temperature: -40 to 65 C (-40 to 149 F)
- Operating humidity: 5% to 93% noncondensing
- Operating altitude: 0 to 3000 m (10,000 ft)
- Non-operating altitude: 12,000 m (40,000 ft)

FAQs

Q: Does this GPU need a licence?
A: No licence is attached to the GPU itself. The host system must be managed in Intersight Managed Mode, which requires a minimum Cisco Intersight Essentials licence per server. NVIDIA vGPU or AI Enterprise software licensing is separate.

Q: Which Cisco systems does it work with?
A: It is designed for the Cisco UCS X440p PCIe Node using Riser A (UCSX-RIS-A-440P), paired with a UCS X210c compute node in a UCS X9508 chassis. Both UCS X9416 X-Fabric modules are required.

Q: What fabric interconnects and software versions are required?
A: Supported Fabric Interconnects are the Cisco UCS 6454, UCS 64108 and UCS 6536. UCS Manager 4.3(4) or later is required, or management through Cisco Intersight.

Q: How does it differ from the UCSX-GPU-H100-NVL?
A: The UCSX-GPU-H100-80 has 80GB of memory and a 350W rating. The UCSX-GPU-H100-NVL is rated at 94GB and 400W.

Q: How is this GPU cooled and powered?
A: It is a passive card with no onboard fan, so it relies on chassis airflow and is powered through the UCS X440p PCIe Node. The node operates at an ambient temperature of 10 to 35 C.

  • 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