How do you give a Kubernetes pod RDMA access to the fabric for NCCL, and what goes wrong when it is done naively?

A pod's default network is an overlay that no RDMA NIC will speak. The three ways to give a pod the fabric (host networking, SR-IOV virtual functions via a NIC device plugin, or a macvlan secondary interface), what each costs in isolation and speed, how NCCL finds the right NIC, and the failure that halves bandwidth.

Updated Sep 2026 · Grounded in real AI infrastructure interview loops and written to a senior-engineer editorial bar, with every number worked and every diagram hand-built.

20 answers per topic instead of 10, plus saved progress and bookmarks · no cardor unlock all 283 remaining answers · ₹2,000 / $25

The concepts behind this question

Ranked by how closely each one overlaps this question's topic, so the first card is the thing to read if the answer above moved too fast.

Foundational

🗂️ Scheduling & Orchestration

Kubernetes GPU SchedulingKubernetes knows nothing about GPUs until something tells it. The NVIDIA device plugin advertises each node's GPUs as a countable resource, the scheduler matches a pod's request to a node with enough of them, and the container runtime wires the device in. That model is enough for one job per GPU and breaks the moment you need sharing, topology or multi-node placement, which is where Dynamic Resource Allocation, the GPU Operator and the batch schedulers come in. Knowing which layer does what is the platform interview's opening question.

Core

🔌 Networking & StorageSign in

RDMA, InfiniBand and RoCEv2Training across nodes moves hundreds of gigabytes per step, and a CPU-driven TCP stack cannot feed a 400 Gb/s link. RDMA lets a NIC write straight into a remote GPU's memory with no kernel and no copies, and it runs over two fabrics: InfiniBand, which is lossless by design, and RoCEv2, which is Ethernet made lossless by configuration. The choice is operational as much as technical, and the numbers that decide it are per-GPU bandwidth, the collective's volume, and who will debug a pause storm at 3 a.m.

Advanced

🔌 Networking & Storage🔒 Premium

GPUDirect RDMA and GPUDirect StorageBy default a byte leaving a GPU for the network or the disk makes a detour through host memory, crossing PCIe twice and costing a CPU copy. GPUDirect RDMA lets the NIC read and write GPU memory directly, and GPUDirect Storage does the same for NVMe. The win is not raw bandwidth (PCIe is the ceiling either way) but the halving of PCIe traffic and the removal of the host as a bottleneck, which is what makes collectives run at NIC rate and checkpoints run at drive rate. When it is silently off, everything still works, at half speed.

Core

🗂️ Scheduling & OrchestrationSign in

MIG, MPS and Time-SlicingA whole H100 is far more than a notebook, a small inference service or a CI job needs, and giving each of them a card leaves most of the fleet idle. Three mechanisms share a GPU, and they differ in what they isolate: MIG partitions the hardware into up to seven slices with their own memory and compute, MPS lets several processes share one GPU's SMs concurrently with no memory isolation, and time-slicing context-switches between processes with no isolation at all. The choice is the isolation the workload needs against the utilization the platform wants.

Explore the full curriculum →

UP NEXT ON YOUR JOURNEY

Next in this trackUpgrade the GPU driver across 2,000 live nodes without breaking running jobs. Walk me through the plan and what can go wrong.Next in this trackA 64-node gang-scheduled job has been pending for six hours while the cluster shows free GPUs. Debug it live.Closest to this topicExplain GPUDirect RDMA. What does it remove from the data path, and when does it change performance?

FEDITOR'S NOTE

Scored on knowing that the pod overlay cannot carry RDMA, on the SR-IOV path (NIC device plugin, VFs, a secondary interface via Multus, RDMA shared or exclusive mode), on GPU-to-NIC affinity, and on the measured check that proves it works.

DISCUSSION · 0

No comments yet — be the first to share your approach.