The Cisco UCS C885A M8 Rack Server is a dense-GPU server designed to deliver massive, scalable accelerated compute capabilities to address the most demanding AI workloads, including deep learning/Large Language Model (LLM) training, model fine-tuning, large model inferencing, and Retrieval-Augmented Generation (RAG).

To deliver massive accelerated compute performance in a single server, the server offers a choice of eight GPUs of the following types:

• NVIDIA^® H100 SXM or NVIDIA® H200 Server PCI Express Module (SXM) GPUs. SXM is a socket-based GPU interconnect method used by NVIDIA GPUs.

• AMD MI300X OCP Accelerator Model (OAM) GPUs. OAM is an Open Compute GPU interconnect standard that avoids GPU vendor lock in.

For north-south traffic, the server also supports one NVIDIA Bluefield-3 B3220 DPU to scale AI model training across a cluster of dense-GPU servers. Up to eight NVIDIA ConnectX-7 or Bluefield-3 B3140H SuperNIC are supported for east-west traffic between GPUs.

The Cisco Baseboard Management Controller (Cisco BMC) web GUI is HTML5 based and added security with SSL (HTTPS). It helps you manage the Cisco UCS C885A M8 Rack Server using the following options:

Hardware and Component Management

The Inventory feature enables administrators to record hardware devices and components on each server, such as central processing units (CPUs), memory modules, hard drives, network cards, and more.

Status and Checks

The Inventory feature also provides status and checks for hardware and software devices. This information can include device health status, temperature, voltage, connection status, and more.