NCCL GPUDirect over RoCE
NCCL GPUDirect over RoCE (RDMA over Converged Ethernet) requires proper configuration for high performance.
NCCL GPUDirect over RoCE (RDMA over Converged Ethernet) requires proper configuration for high performance.
What this failure is
NCCL GPUDirect over RoCE is a Communication failure seen during ML training runs. NCCL GPUDirect over RoCE (RDMA over Converged Ethernet) requires proper configuration for high performance. Common tags: Roce, Rdma, Nccl, Gpudirect.
Is this what broke your run? Paste your log.
You're reading about NCCL GPUDirect over RoCE. Paste your own crash log or traceback below and get the real root cause for YOUR run, not this generic entry. No account, no card. Logs are masked at ingress and never saved to account history.
Want 14 days on the Scale plan?
Request an evaluation code. A verified workplace organization activates up to 50 diagnoses a day, alerts, history, and follow-up questions. No credit card or automatic subscription.
Why it happens (the mechanism)
RoCE not enabled on the NIC. GPUDirect RDMA not configured for RoCE. PFC and ECN not configured on switches. NCCL not configured for RoCE transport. Taken together, these mechanisms explain why the failure is reproducible, why it tends to surface on specific workloads or scales, and why generic mitigation attempts often fall short without addressing the underlying cause.
What you'll observe
- Multi-node training is slow on Ethernet fabric
- GPUDirect RDMA not working on RoCE
- NCCL performance is poor on RoCE
Common symptoms and what they mean
| Symptom | Why it happens |
|---|---|
| NCCL WARN RoCE is not enabled | RoCE not enabled on the NIC |
| GPUDirect RDMA is not available on RoCE | GPUDirect RDMA not configured for RoCE |
| NCCL performance is 10x slower than expected | PFC and ECN not configured on switches |
Which systems are affected
- Multi-node training on Ethernet with RDMA-capable NICs
- Cloud GPU instances with RoCE support
- GPU clusters with Mellanox/NVIDIA ConnectX NICs
How to confirm this is the problem
Use this checklist to test the hypothesis against a small reproduction. No single line proves the root cause, so preserve the preceding events and compare one variable at a time.
- ✓Reproduce the failure from a clean checkpoint/seed: the symptom must appear without warm-up state from a previous run.
- ✓Verified signal present: NCCL WARN RoCE is not enabled
- ✓Verified signal present: GPUDirect RDMA is not available on RoCE
- ✓Verified signal present: NCCL performance is 10x slower than expected
- ✓A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.
The fix and the prevention pattern
The root cause is on this page and stays free. A free account adds the exact remediation steps, saved history, and the fix on every entry in the encyclopedia.
Sign up free. Unlock the full analysisNo credit card. Daily allowance follows verified trust tier. Instant access.
Diagnose this failure in VS Code
Select the traceback or open the failed terminal, then run Denpex locally to see the initiating rank, collateral failures, exact fix, and verification command without uploading the log.
Install the free VS Code extensionNCCL errors in context
NCCL is where a distributed job reports failure, which is not the same as where it failed. The hub lists every common NCCL error next to what it actually indicates, and the environment variables that tell them apart.
Compare every nccl error side by sideRelated failures to investigate next
Root cause
- RoCE not enabled on the NIC
- GPUDirect RDMA not configured for RoCE
- PFC and ECN not configured on switches
- NCCL not configured for RoCE transport
The fix and how to prevent it
Evaluate Denpex on your own logs
Request a Scale evaluation code. A verified workplace organization activates 14 days with up to 50 diagnoses a day. Every account keeps its current diagnosis allowance and gets a verification path. No card or automatic subscription.
Don't just read the fix, diagnose your run
The encyclopedia tells you what went wrong. Denpex tells you what went wrong in YOUR training run. With your logs, your config, and your stack.