Getting RDMA into Kubernetes Pods, Part 1: hostNetwork

Thu, 12 Nov 2026 00:00:00 +0000

There are four common ways to give a Kubernetes pod RDMA for GPU training:

The pod shares the network of the node with hostNetwork.
The pod gets an SR-IOV virtual function (VF) on each rail.
The pod gets the RDMA devices of the rails and an IPvlan interface on each rail.
A DRA driver gives the pod its own interface on each rail.

This series tests all four on the same two A100 nodes, so the results can be compared.

Gpu on Under the Model