Skip to product information
1 of 1

Cisco

Cisco UCSX-GPU-A16 | UCS X-Series NVIDIA A16 GPU, 4x16GB, 250W, PCIe Gen4

SKU:UCSX-GPU-A16

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 We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 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 UCS X-Series NVIDIA A16 is a dual-slot, full-height, full-length PCIe Gen4 GPU accelerator built for high-density virtual desktop (VDI) deployments. It is a GPU option for the Cisco UCS X440p PCIe node, which pairs with UCS X210c compute nodes in the UCS X9508 chassis. Typical uses are VDI with high user density under NVIDIA Virtual PC (vPC), along with video transcoding and Android cloud gaming.

Each card carries four NVIDIA Ampere GA107 GPUs with 16 GB of GDDR6 per GPU, 64 GB in total. Maximum board power is 250 W (235 W MaxQ, 195 W minimum), with a passive thermal design that relies on system airflow. The interface is physical x16 and runs at PCIe Gen4 x16, with Gen3 x16 also supported. This SKU is the NVIDIA A16, 250W, 4x16GB variant supplied with a power cable, and each GPU supports up to 16 SR-IOV virtual functions.

The card installs in the UCS X440p PCIe Node (Riser A, one dual-slot GPU per riser) and requires UCS X9416 X-Fabric modules with UCS 6454, 64108 or 6536 Fabric Interconnects. It is managed through Cisco Intersight or UCS Manager 4.3(4) or later. NVIDIA vPC, vApps, RTX vWS or vCS software licences are required for virtualised use and are purchased separately.

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

Features

- Four GPUs on one board give high encoder, decoder and frame-buffer density in a dual-slot form factor for VDI
- Based on the NVIDIA Ampere architecture, aimed at NVIDIA Virtual PC (vPC) software
- Also suited to video transcoding and Android cloud gaming workloads
- Passive thermal design handles challenging ambient environments (NEBS-3 capable)
- Supports SR-IOV, MSI-X and ARI forwarding
- Secure Boot and Root of Trust supported

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: 4 x NVIDIA GA107-890 (Ampere)
- Memory size: 64 GB (16 GB per GPU)
- Memory type: GDDR6
- Memory clock: 6.25 GHz
- Memory bus width: 128 bits (per GPU)
- Peak memory bandwidth: Up to 4x 200 GB/sec
- GPU clocks: Base 1312 MHz, Boost 1755 MHz
- Total board power: 250 W default, 250 W maximum, 235 W MaxQ, 195 W minimum
- Thermal solution: Passive (requires system airflow)
- Form factor: Full-height, full-length (FHFL) 10.5 in, dual-slot
- PCI Express interface: Physical x16; PCIe Gen4 x16 or Gen3 x16 supported
- SR-IOV support: 16 virtual functions per GPU
- Zero Power: Not supported
- ECC support: Enabled by default (can be disabled via software)
- Driver support: R470 or later
- CUDA support: CUDA 11.4 or later
- Virtual GPU software support: vGPU 13.0 or later (NVIDIA vPC, vApps, RTX vWS, vCS)
- Board weight: 1088 g (excluding bracket and extenders)
- Ambient operating temperature: 0 to 50 C
- Storage temperature: -40 to 75 C
- Operating humidity: 5% to 85% relative humidity
- Storage humidity: 5% to 95% relative humidity
- MTBF: 677,793 hours at 35 C (uncontrolled environment); 874,410 hours at 35 C (controlled environment)

FAQs

Q: Does this GPU include NVIDIA vGPU software licensing?
A: No. The GPU hardware does not include NVIDIA vGPU software licensing. NVIDIA Virtual PC (vPC), vApps, RTX vWS or vCS licences are required for virtualised use and are purchased separately, using vGPU 13.0 or later. Intersight Managed Mode also requires a minimum Essentials licence per server.

Q: How much power does the A16 need?
A: Maximum board power is 250 W, with 235 W MaxQ and 195 W minimum. This SKU is the variant supplied with a power cable, and NVIDIA documents a CPU 8-pin to PCIe 8-pin power adapter for the card.

Q: Which Cisco systems does it install in?
A: It installs in the Cisco UCS X440p PCIe Node (Riser A, one dual-slot GPU per riser), which pairs with UCS X210c compute nodes in the UCS X9508 chassis. It requires UCS X9416 X-Fabric modules and UCS 6454, 64108 or 6536 Fabric Interconnects.

Q: Which management software versions are supported?
A: The card is managed through Cisco Intersight or UCS Manager 4.3(4) or later. Driver support is R470 or later, and CUDA 11.4 or later is supported.

Q: What workloads is the A16 designed for?
A: It is built for high-density VDI using NVIDIA vPC, with four GPUs on one board providing high encoder, decoder and frame-buffer density. It also suits video transcoding and Android cloud gaming.

  • 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