Skip to product information
1 of 1

Cisco

Cisco C885A-M8-CC-SLD18 | UCS C885A M8 CPU Sled, EPYC 9575F, 96GB DDR5, CX-7

SKU:C885A-M8-CC-SLD18

Request Quote
Sale Sold out
Shipping calculated at checkout.

Stock Status: Enquire

20-min response

Need a quote or bulk pricing? Answered within 20 minutes, Sydney business hours.

Description

This CPU sled is a compute module for the Cisco UCS C885A M8, an 8RU dense-GPU rack server built for AI training, model fine-tuning, large-model inferencing and RAG workloads. It suits data centre teams building out GPU-heavy AI and HPC infrastructure on the UCS C885A M8 platform.

This variant is built around the high-frequency AMD EPYC 9575F processor with 96GB of DDR5 memory and integrates NVIDIA ConnectX-7 network adapters for scale-out AI networking. The host server supports two-socket AMD EPYC processors, up to 24 DDR5-6400 MT/s DIMM slots, and 8x PCIe Gen5 x16 HHHL slots for east-west traffic, with 5x PCIe Gen5 x16 FHHL slots for north-south traffic. The CPU tray is a 3RU chassis within the 8RU main server chassis and is powered by 2x 2700W 12V CRPS Titanium supplies.

The sled installs in the Cisco UCS C885A M8 Rack Server, which supports 8x AMD MI350X OAM GPUs and is managed through Cisco Intersight. Recommended optics for the server include QSFP-200G-SL4, QSFP-200G-SR4-S, QSFP-400G-DR4 and QSFP-200G-FR4-S, used with MPO12 trunk cables.

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

Features

- Compute module for the Cisco UCS C885A M8 dense-GPU AI server
- High-frequency AMD EPYC 9575F CPU paired with DDR5 memory
- Integrated NVIDIA ConnectX-7 NICs for scale-out AI networking
- Host server designed for AI training, fine-tuning, large-model inferencing and RAG workloads
- Host server supports 8x AMD MI350X OAM GPUs
- Managed through Cisco Intersight
- Redundant 1+1 Titanium-rated power for the CPU tray

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

- CPU: AMD EPYC 9575F
- Memory: 96GB DDR5
- Networking: NVIDIA ConnectX-7 NICs (CX-7s)
- Host server platform: Cisco UCS C885A M8 Rack Server, 8RU
- CPU tray form factor: 3RU CPU chassis within 8RU main chassis
- Server CPU support: two-socket 5th Gen or 4th Gen AMD EPYC processors
- Server memory: up to 24 DDR5-6400 MT/s DIMM slots, up to 3 TB (24x 128GB)
- Server networking slots: 8x PCIe Gen5 x16 HHHL (East-West), 5x PCIe Gen5 x16 FHHL (North-South), 2x OCP 3.0 SFF
- Server boot storage: up to 2x PCIe3 x4 960 GB M.2 NVMe
- Server data storage: up to 16x PCIe5 x4 2.5 inch U.2 NVMe SSD
- Server management: DC-SCM with RJ45, IPMI 2.0, KVM, dedicated LAN, watchdog
- Server dimensions (H x W x D): 353 x 447 x 800 mm
- Server operating temperature: 5 to 35 C
- Server operating humidity: 8% to 90% non-condensing
- Server input power: 200-240 VAC, 50-60 Hz, 16A per PSU
- CPU tray power: 2x 2700W CRPS @12V (1+1), 80 PLUS Titanium

FAQs

Q: Which server does this CPU sled fit?
A: It installs in the Cisco UCS C885A M8 Rack Server, an 8RU dense-GPU platform for AI and HPC workloads.

Q: Does the sled need its own licence?
A: No, the sled itself has no separate licence. The host server is managed through Cisco Intersight, which requires an Infrastructure Services Essentials or Advantage tier.

Q: What power does the CPU tray need?
A: The CPU tray is powered by 2x 2700W 12V CRPS Titanium supplies in a 1+1 configuration, running from 200-240 VAC at 50-60 Hz.

Q: Which optics are recommended for the server?
A: Recommended optics are QSFP-200G-SL4, QSFP-200G-SR4-S, QSFP-400G-DR4 and QSFP-200G-FR4-S, used with MPO12 trunk cables.

Q: How does this sled differ from other C885A M8 CPU sleds?
A: This variant uses the AMD EPYC 9575F processor with 96GB DDR5 and NVIDIA ConnectX-7 network adapters. Other sleds in the range differ in CPU, memory and NIC configuration.

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