Skip to product information
1 of 1

Cisco

Cisco UCSC-885A-M8-H25 | UCS C885A M8 8x H200 GPU Server, 3TB, 8x 400GbE

SKU:UCSC-885A-M8-H25

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 We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 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 dense-GPU rack server built for demanding AI workloads, including LLM training, model fine-tuning, large-model inferencing and Retrieval-Augmented Generation. It also suits HPC and scale-out AI clusters made up of GPU-dense servers.

This H25 configuration pairs eight NVIDIA H200 Tensor Core SXM GPUs on a UBB sled with two AMD EPYC 9575F CPUs and 3TB of DDR5-6400 memory (24 x 128GB DIMMs). East-West connectivity comes from eight NVIDIA ConnectX-7 single-port 400GbE NICs, one per GPU, while North-South traffic runs over two ConnectX-7 dual-port 200GbE NICs. An OCP Intel X710 T2L NIC and two 1.92TB NVMe SSDs complete the build. Power is supplied by six 3000W 54V AC supplies in a 4+2 arrangement plus two 2700W 12V Titanium supplies, with a maximum system draw of about 12.5 kW.

The C885A M8 platform also supports other GPU and NIC options, but this SKU is a fixed H200 and ConnectX-7 configuration.

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

Features

- Supports up to eight GPUs per system for AI training, fine-tuning, inference and RAG
- NVIDIA H200 SXM GPUs on a UBB (HGX-style) sled
- Eight 400GbE ConnectX-7 NICs, one per GPU, for scale-out AI cluster fabrics
- Two dual-port 200GbE ConnectX-7 NICs for North-South connectivity
- Hot-swappable NVMe and SAS SSD front bays
- Redundant 54V power supplies with 80 PLUS Titanium efficiency
- Separate CPU tray and GPU tray with front-mounted fans
- Fixed, pre-validated configuration optimised for AI and HPC

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

- GPUs: 8 x NVIDIA H200 Tensor Core SXM (UBB GPU sled, C885A-M8-H2SX-SLD)
- CPUs: 2 x AMD 9575F (CPU sled C885A-M8-CC-SLD12)
- Memory: 3TB (24 x 128GB DDR5-6400 DIMMs)
- East-West NICs: 8 x NVIDIA ConnectX-7 single-port 400GbE
- North-South NICs: 2 x NVIDIA ConnectX-7 dual-port 200GbE
- OCP NIC: 1 x Intel X710 T2L
- Drives: 2 x 1.92TB NVMe SSD
- Power supplies: 6 x 3000W 54V AC plus 2 x 2700W 12V Titanium
- Power redundancy: 4+2 on the 54V supplies, 80 PLUS Titanium
- Maximum system power: approximately 12.5 kW
- AC input: 200-240 VAC, 50-60 Hz, 16A per PSU
- Front I/O: 2 x USB 2.0, power button with LED, UID button with LED
- Height: 353 mm (13.9 in)
- Width: 447 mm (17.6 in)
- Depth: 800 mm (31.5 in)
- Weight: approximately 120 kg (265 lb)
- Operating temperature: 5 to 35 C
- Operating humidity: 8% to 90%, non-condensing
- Emissions: FCC Class A, CISPR 32 / EN55032, AS/NZS CISPR 32

FAQs

Q: What power does this server need?
A: It uses six 3000W 54V AC supplies in a 4+2 redundant arrangement plus two 2700W 12V Titanium supplies. Maximum system power is about 12.5 kW, and each supply needs a 200-240 VAC, 50-60 Hz feed rated for 16A.

Q: How is networking arranged on this configuration?
A: East-West traffic uses eight single-port 400GbE ConnectX-7 NICs, one per GPU. North-South traffic uses two dual-port 200GbE ConnectX-7 NICs, and an OCP Intel X710 T2L NIC is also fitted.

Q: Can I choose different GPUs in this SKU?
A: No. UCSC-885A-M8-H25 is a fixed configuration with H200 GPUs and ConnectX-7 NICs. The wider C885A M8 platform also supports NVIDIA H100 and AMD Instinct MI300X/MI350X GPUs in other configurations.

Q: What workloads is it designed for?
A: It targets deep learning and LLM training, model fine-tuning, large-model inferencing, Retrieval-Augmented Generation and HPC, particularly in scale-out clusters of dense-GPU servers.

Q: What are the physical and environmental requirements?
A: The chassis measures 353 mm high, 447 mm wide and 800 mm deep, and weighs approximately 120 kg. Operating temperature is 5 to 35 C with 8% to 90% non-condensing humidity.

  • 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