Is Ethernet ready for AI? The proof is shipping

The hardware is here

Open 800G AI switches are not a preview. They are on the hardware compatibility list and shipping. OcNOS-DC runs on Broadcom Tomahawk 5 and Tomahawk 4 platforms from Edgecore and UfiSpace, the same merchant silicon the rest of the industry builds AI fabrics on, so you keep switch, NIC, and GPU choices open. These are on-chip, shared-buffer switches built for lossless RDMA at 800G.

The scale is proven

The most common objection to Ethernet for AI is scale, and the industry has already answered it in production. You do not have to take a switch vendor's word for it, because the hyperscalers publish their designs.

On OcNOS-DC the same physics apply to your build. A two-tier Tomahawk 5 design reaches 4,096 GPUs at 800G per GPU and a 1:1 subscription ratio. Above that, a super-spine tier extends to 16,000 or more. See the AI fabric topologies for the port math behind each tier.

The utilization gap is closed

The old knock on Ethernet was load balancing: standard ECMP pins each large flow to one uplink, so during a collective some links jam while parallel paths sit idle and the sync waits on the slowest link. That gap is now closed on the same switches, and the proof is a hardware feature, not a promise.

The standard is moving

Ethernet for AI is not a single vendor's proprietary detour. The Ultra Ethernet Consortium published the UEC 1.0 specification to adapt Ethernet transport for AI and HPC scale, backed by the companies that build the silicon, the NICs, and the clouds. IP Infusion is a member of the Ultra Ethernet Consortium.

UEC 1.0 defines a multipath transport that spreads one flow across many paths instead of pinning it to a single ECMP hash, with that endpoint behavior running on UEC-capable NICs. OcNOS-DC runs lossless RoCEv2 with PFC, ECN marking, and DLB in production today on Broadcom Tomahawk 4 and Tomahawk 5 switches. DCQCN runs on the NICs and reacts to those ECN marks. See the Ultra Ethernet deep-dive for how packet spray and multi-path RDMA change the transport.

What OcNOS-DC ships today

Underneath the proof points is a fabric you can turn up now. OcNOS-DC provides the lossless Ethernet fabric that RoCEv2 requires, so your training collectives move GPU-to-GPU over RDMA without loss. DCQCN is the end-to-end loop between the NICs and the switches; the switch role in that loop is early ECN marking. Every primitive below is verified and shipping, not a roadmap item.

What is coming next

The items below are roadmap and upcoming-silicon, labeled as such. They are not shipping capabilities today, and none of the figures are IP Infusion performance numbers.