Deploy and version the NVIDIA operator stack correctly, push RoCE and GPUDirect settings into NIC firmware safely, and align NIC to GPU so a training job runs at full speed.

1. 3.1 Network Operator and the NicClusterPolicyThe policy that was applied and never became anythingAp
2. 3.2 Version matrices, upgrades and heterogeneous clustersThe bookmark that says 1.31 is fineE
3. 3.3 NIC firmware config in Kubernetes: RoCE, PFC and rebootsThe change request that says no outage expectedAn
4. 3.4 GPUDirect RDMA: dma-buf, nvidia-peermem and GDSlsmod says nothing, so GPUDirect must be brokenAn
5. 3.5 NUMA alignment with Topology Manager and CPU ManagerSix nodes are slower and nobody can say whichE
6. Checkpoint