Cisco UCS C885A M8 (HGX H100/H200)
8 × H100 / H200 SXM per node, with 8 compute rails at 400G. That is 3.2 Tb/s of scale-out network per node.
- GPUs per node8 × H100 / H200 SXM
- Compute rails8 × 400G
- Scale-out bandwidth3.2 Tb/s
- Storage links2 × 400G
- In-band links2 × 100G
- BMC ports1
- Rack units8U
- Typical power10.2 kW
- Maximum power10.5 kW
What this means for the fabric
8 compute NICs means 8 rails. In a rail-optimized design each of those rails is sized on its own against the node count, then multiplied back up, so the leaf tier is driven by the rail count and not by the total endpoint count.
Storage and in-band management are separate fabrics with their own switches and their own optics, and the 1 BMC port per node feed the out-of-band network, which also has to carry every switch and every PDU in the cluster.
Power and racks
At 10.2 kW typical draw, this node fills a rack's power budget long before it fills the space. A 45 kW rack takes 4 of them on power alone, against 5 on rack units. Power is what actually sets the rack count.
Notes
Cisco's first eight-way HGX system. Verified: 8U 19-inch EIA, up to six 3 kW supplies in N+1, 8 x HGX H100 or H200 (also available with AMD MI300X/MI350X), two 4th or 5th Gen AMD EPYC, and one ConnectX-7 or BlueField-3 SuperNIC per GPU plus BlueField-3 DPUs for storage. Network-identical to a DGX H100, so fabric sizing matches.
Source: Cisco UCS C885A M8 data sheet and at-a-glance, verified 2026-08. Rack units, GPU count, and per-GPU networking confirmed; power figures are estimates. Needs verification 2026-08-06