Skip to product information
1 of 1

Cisco

Cisco UCSC-885A-M8-H29 | UCS C885A M8 H200 GPU Server, 12x 400GbE, 3TB

SKU:UCSC-885A-M8-H29

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 We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 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 C885A M8 is a fixed-configuration 8RU dense-GPU rack server built on the NVIDIA HGX platform. This H29 configuration pairs NVIDIA H200 GPUs with twelve 400GbE NVIDIA ConnectX-7 NICs and 3TB of memory, aimed at large language model training, fine-tuning, large-model inferencing and Retrieval-Augmented Generation in AI and HPC clusters.

The two-socket platform uses 5th Gen AMD EPYC processors, with up to 24 DDR5-6400 DIMM slots. Local storage supports up to 2 x 960GB M.2 NVMe boot drives and up to 16 PCIe Gen5 x4 2.5-inch U.2 NVMe SSDs for data caching. Networking includes 8 PCIe Gen5 x16 slots for East-West traffic and 5 PCIe Gen5 x16 slots for North-South traffic, plus an OCP 3.0 slot with 2x 10GbE RJ-45. Power comes from 6 x 3000W 54V hot-swap supplies (N+2) and 2 x 2700W 12V supplies (N+1), at up to 12.5 kW system draw on 200-240 VAC input.

The East-West slots accept NVIDIA ConnectX-7, Pollara or BlueField-3 B3140H NICs. The server is managed through Cisco Intersight, and each product ID is a single fixed configuration.

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

Features

- Built on the NVIDIA HGX platform for dense-GPU AI and HPC workloads
- One 400G NIC per GPU for East-West scale-out training across GPU server clusters
- Two 5th Gen AMD EPYC CPUs on the platform
- Hot-swappable, redundant power supplies for high availability
- Local NVMe storage for data caching
- Managed through Cisco Intersight
- Hot-swappable N+1 cooling fans

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

- Chassis: 8RU rack server
- Platform: NVIDIA HGX
- GPU: NVIDIA H200 (8-GPU HGX board)
- Network adapters: 12 x 400GbE NVIDIA ConnectX-7
- System memory: 3TB
- Memory capacity (platform): up to 24 DDR5-6400 DIMM slots
- CPU (platform): two-socket 5th Gen or 4th Gen AMD EPYC
- East-West slots: 8 PCIe Gen5 x16 HHHL
- North-South slots: 5 PCIe Gen5 x16 FHHL
- OCP: 2 OCP 3.0 SFF slots, 1x NIC with 2x 10GbE RJ-45
- Boot storage: up to 2 x 960GB PCIe3 x4 M.2 NVMe
- Data cache storage: up to 16 x PCIe5 x4 2.5-inch U.2 NVMe SSD
- Power supplies: 6 x 3000W 54V (N+2) and 2 x 2700W 12V (N+1)
- System power consumption: up to 12.5 kW
- Input voltage: 200-240 VAC, 50-60 Hz
- Cooling: 12 x 80105 hot-swappable fans (N+1), 4 x 6056 SSD cooling fans
- Management: host BMC via DC-SCM, RJ45, IPMI 2.0, KVM with dedicated LAN
- Dimensions (H x W x D): 353 x 447 x 800 mm
- Weight: 120 kg
- Operating temperature: 5 to 35 C

FAQs

Q: Does this server need a licence or subscription?
A: Cisco Intersight is a required selection in the configuration guide for management. An NVIDIA AI Enterprise Essentials subscription, licensed per GPU, is optional.

Q: What power does it require?
A: It runs on 200-240 VAC input and draws up to 12.5 kW, using 6 x 3000W 54V supplies (N+2) and 2 x 2700W 12V supplies (N+1). Input power cords are selected separately.

Q: Which NICs does the East-West network support?
A: The 8 PCIe Gen5 x16 East-West slots support NVIDIA ConnectX-7, Pollara and BlueField-3 B3140H NICs or SuperNICs.

Q: Can the GPU, NIC or memory configuration be changed?
A: No. The UCS C885A M8 is offered in fixed configurations, and each product ID is a single configuration. This SKU is the H200 GPU, 12x 400GbE ConnectX-7, 3TB memory build.

Q: What workloads is it designed for?
A: It is built for large deep learning and LLM training, model fine-tuning, large-model inferencing and Retrieval-Augmented Generation in AI and HPC clusters.

  • 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