CollectiveX: experimental cross-vendor collective/EP benchmark#1896
CollectiveX: experimental cross-vendor collective/EP benchmark#1896Oseltamivir wants to merge 16 commits into
Conversation
Per-SKU launch adapters (launch_<sku>.sh) that run any benchmark via a CX_BENCH selector through a shared run_in_container.sh; multi-arch digest-pinned sglang container; NCCL-primitive + DeepEP dispatch/combine benchmarks with provenance + correctness gating; and an on:push workflow (GB200 NCCL smoke; workflow_dispatch for B200/DeepEP/larger sweeps). Validated on hardware: NCCL primitives on B200 (8x NVLink) and GB200 (4x NVL72 MNNVL); DeepEP dispatch/combine on GB200 (correctness-gated).
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
![github-code-quality[bot]](https://avatars.githubusercontent.com/u/9919?s=60&v=4){kind=small}
The GB200 on:push smoke hung 25 min in enroot import: a bare digest ref (repo@sha256:) can't form an anonymous Docker Hub token scope, so enroot prompted for a password and blocked in non-interactive CI. Import by the multi-arch TAG instead (anonymous auth works, same as the serving launchers) and add </dev/null so a missing token fails fast rather than hanging. Use v0.5.11-cu130 (multi-arch amd64+arm64, index sha256:061fb71f…): v0.5.12-cu130's 62 layers overflow enroot's overlay-based squash creation on these nodes (failed to mount overlay … Invalid argument). v0.5.11-cu130 imports cleanly and is pre-staged on GB200.
On the GB200 Actions path, CX_STAGE_DIR makes the launcher rsync the tree to compute-visible Lustre and the container writes results/ there; upload-artifact reads the checkout's results/ (empty), so the green smoke produced no artifact. Add cx_collect_results to copy result JSONs from the stage dir back to the checkout after the run (no-op when no staging was used).
Add summarize.py (compact NCCL/DeepEP results table, printed at end of every job) and make it the result gate. Fix review findings: benchmark failures/skipped-deepep now fail the job instead of reporting green (#1); DeepEP nodes from SLURM_NNODES not world_size//8 (#3); apply Buffer.set_num_sms so num_comm_sms is real (#8); nccl-tests -c 1 with a missing check footer is now invalid (#7); use context managers for file reads (#4,#5); launchers export COLLECTIVEX_IMAGE/_DIGEST for provenance (#9); trim workflow_dispatch sku options to launcher-backed pools (#2). Artifact-path finding (#6) already fixed via cx_collect_results.
| is_token_in_rank=is_token_in_rank, | ||
| num_tokens_per_expert=num_tokens_per_expert, | ||
| ) | ||
| combined_x, _, _ = buffer.combine(recv_x, handle, topk_weights=recv_topk_weights) |
There was a problem hiding this comment.
Dispatch dtype not applied
Medium Severity
The --dispatch-dtype / CX_DISPATCH_DTYPE value is stored in result metadata but never used when building inputs or calling buffer.dispatch. Runs always use bfloat16 token tensors regardless of fp8 vs bf16, so provenance and comparison keys can describe a different shape than what was measured.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit b384171. Configure here.
summarize.py --markdown emits GitHub-flavored markdown tables (NCCL + DeepEP); a per-job 'Results summary' workflow step appends it to $GITHUB_STEP_SUMMARY so the run page shows a rendered table (per the GitHub job-summaries feature). Plain-text mode still drives the in-container result gate.
| --timestamp "$TS" || cx_log "WARN: parse $op failed" | ||
| done | ||
|
|
||
| cx_log "done — JSON artifacts under $CX_DIR/results/" |
There was a problem hiding this comment.
Multinode launcher ignores failures
High Severity
The B200 multinode adapter logs warnings when srun or run_nccl.py fail but always exits successfully. Unlike run_in_container.sh, it never runs summarize.py as a non-zero gate, so workflow_dispatch on b200-multinode can finish green with no valid NCCL results.
Reviewed by Cursor Bugbot for commit f48daed. Configure here.
| run: bash "experimental/CollectiveX/launchers/launch_${RUNNER_NAME%%_*}.sh" | ||
| - name: Results summary | ||
| if: always() | ||
| run: python3 experimental/CollectiveX/summarize.py --results-dir experimental/CollectiveX/results --markdown >> "$GITHUB_STEP_SUMMARY" |
There was a problem hiding this comment.
Workflow skips result failure gate
Medium Severity
Both jobs only run summarize.py --markdown, which is documented to always exit 0. The workflow never runs the plain summarize.py gate on the checkout’s results/ after launch, so a successful Launch step can stay green when the checkout has no valid JSON (e.g. staged runs where copy-back failed).
Additional Locations (1)
Reviewed by Cursor Bugbot for commit f48daed. Configure here.
| dst="$repo_root/experimental/CollectiveX/results" | ||
| mkdir -p "$dst" | ||
| cp "$mount_src/experimental/CollectiveX/results/"*.json "$dst/" 2>/dev/null || true | ||
| cx_log "copied results from stage dir -> $dst (for artifact upload)" |
There was a problem hiding this comment.
Result copy errors ignored
Medium Severity
cx_collect_results wraps the staged-to-checkout cp in 2>/dev/null || true and always logs success, so a failed or empty copy does not affect the launcher exit code and the workflow can pass without uploadable JSON.
Reviewed by Cursor Bugbot for commit f48daed. Configure here.
First AMD / cross-vendor reach, scaffolded ahead of Milestone 1: - run_mori.py: MoRI dispatch+combine (normal mode), correctness-gated, mirroring ROCm/mori's dispatch_combine example — int32 routing indices, (n,0) fp8 scales, the zero-copy registered-combine-input-buffer staging step, and expected = input x (#unique destination ranks). Emits the same flat JSON shape (family=moe, backend=mori) with CUDA-event timing. - launchers/launch_mi355x-amds.sh: AMD adapter — partition compute, no account, --cpus-per-task=128, node-local /var/lib/squash imported via srun on the allocated node, --container-writable --container-remap-root, forces CX_BENCH=mori, mounts the (compute-visible) checkout at /ix. - launchers/run_in_container.sh: run_mori_suite + mori case (nccl|deepep|mori|all). - launchers/common.sh: ROCm MoRI image (rocm/sgl-dev:...-mori-0227-2) in cx_default_image for mi355x*/mi350x*/mi325x*/mi300x*. - workflow: mi355x sku + mori benchmark options for workflow_dispatch. - docs: CONTAINERS.md AMD section, README files/run/risks, plan.md status. Not yet hardware-validated (no MI355X access) — MoRI's Python API is version-sensitive (marked ADAPT HERE); the first runner job is the validation, as GB200 was for DeepEP. The ROCm image isn't digest-pinned yet.
- workflow: replace the on:push GB200 NCCL smoke with the MI355X MoRI dispatch/combine run (runs-on: mi355x, CX_BENCH=mori), and name the job "CollectiveX Experimental" (no longer "smoke"). GB200/B200 NCCL + DeepEP remain on workflow_dispatch. - launch_mi355x-amds.sh: adapt more faithfully to runners/launch_mi355x-amds.sh — squeue by job-name only (no -u), flock -w 600, and clear ROCm gpucore.* dumps after the run so the next checkout is clean. Bump default CX_TIME to 60 for a cold ROCm-image import. - summarize.py: drop the "N/N results valid." footer from both the job-summary (markdown) and plain output; the failure gate still reports invalid results. Relabel the MoE section "MoE dispatch+combine (DeepEP / MoRI)". - docs: README/plan describe push -> MI355X MoRI.
| rm -f \"$SQUASH_FILE\" | ||
| enroot import -o \"$SQUASH_FILE\" \"docker://$IMAGE\" </dev/null | ||
| fi | ||
| " |
There was a problem hiding this comment.
MI355X import errors ignored
High Severity
The node-local enroot import runs inside an srun bash snippet without set -e and with no check after import. A failed import still yields exit 0 from that snippet, so the job continues into pyxis with a missing or corrupt squash file.
Reviewed by Cursor Bugbot for commit d8ee9bf. Configure here.
| - name: Launch ${{ inputs.sku }} / ${{ inputs.benchmark }} | ||
| env: | ||
| RUNNER_NAME: ${{ runner.name }} | ||
| run: bash "experimental/CollectiveX/launchers/launch_${RUNNER_NAME%%_*}.sh" |
There was a problem hiding this comment.
Workflow skips multinode staging
Medium Severity
CX_STAGE_DIR is set only when inputs.sku is gb200. The b200-multinode dispatch target uses launch_b200-dgxc-slurm.sh, which documents the same compute-visible checkout requirement but leaves staging unset, so Slurm jobs may not see the repo mount.
Reviewed by Cursor Bugbot for commit d8ee9bf. Configure here.
… default) First MI355X run reached the MoRI dispatch kernel — salloc, ROCm-image import, mount, torchrun, 8-rank Gloo + shmem init, and EpDispatchCombineConfig/op/dispatch all worked, confirming the API signatures. It OOM'd MoRI's default 2 GiB static symmetric heap (hidden=7168 dispatch/combine buffers across 8 ranks request ~0.9 GiB each). run_mori.py now sets MORI_SHMEM_HEAP_SIZE before `import mori` (default 16 GiB, override CX_MORI_HEAP_BYTES). Docstring + CONTAINERS.md record the finding; correctness/timing validated by the heap-sized re-run.
|
|
||
| salloc --partition="$PARTITION" --exclude="$EXCLUDE_NODES" --gres=gpu:"$NGPUS" \ | ||
| --exclusive --cpus-per-task=128 --time="$TIME_MIN" --no-shell --job-name="$RUNNER_NAME" | ||
| JOB_ID="$(squeue --name="$RUNNER_NAME" -h -o %A | head -n1)" |
There was a problem hiding this comment.
Slurm job ID not scoped
Medium Severity
launch_mi355x-amds.sh resolves JOB_ID with squeue --name="$RUNNER_NAME" and no -u "$USER", while the other CollectiveX NVIDIA launchers filter by user. On a shared cluster, the first matching job name may belong to another account, so subsequent srun/scancel can target the wrong allocation.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit ac3f1b9. Configure here.
The heap-bump run cleared the 2 GiB OOM but then failed registering the 16 GiB symmetric heap as an RDMA memory region (errno 22 EINVAL, size=17179869184). ROCm/mori's reference test uses MORI_SHMEM_HEAP_SIZE="6G" single-node — big enough for the hidden=7168 dispatch/combine buffers, small enough to register. Match it: default "6G" (override CX_MORI_HEAP_SIZE). The rest of the config already matches the reference (max_num_inp_token_per_rank=4096, hidden=7168, backend cpu:gloo,cuda:nccl), so this lands on the proven single-node setup.
Drove run_mori.py to a correct run on 8x MI355X (on-node via salloc+srun): dispatch+combine numerically correct (combine within tol, max_rel ~2e-3), ~85us round-trip at the decode shape. The first runs surfaced four issues, all fixed and re-validated: - RDMA MR ceiling: MoRI registers the WHOLE symmetric heap as one RDMA MR at init (even single-node; no disable-RDMA knob). The ionic_rdma NICs cap GPU MRs at ~4 GiB — a 6 GiB heap fails (RegisterRdmaMemoryRegion errno 22), 2 GiB registers. Hold heap at MORI_SHMEM_HEAP_SIZE=2G (override CX_MORI_HEAP_SIZE). - Buffer sizing: max_num_inp_token_per_rank 4096 -> max(512, n) so the buffers fit the 2 GiB heap (4096 was inherited from the reference test). - Correctness shape: combine returns the full max-token buffer; compare only combined[:n] against expected. - recv count: read total_recv BEFORE combine (combine resets recv_num, which made recv_nonzero a false negative). - Teardown: MoRI's shmem teardown asserts (CheckStatusValid -> SIGABRT) when the op is destroyed after shmem_finalize(); hard-exit after writing results. Docs (README/plan/CONTAINERS) updated from "scaffolded" to validated, with the fabric constraints recorded.
| # the ranks and hard-exit, skipping the buggy finalize/destructor path. | ||
| try: | ||
| dist.barrier() | ||
| except Exception: |
| return hashlib.sha256("|".join(parts).encode()).hexdigest()[:16] | ||
|
|
||
|
|
||
| def main() -> int: |
…CH=nccl) Adds the AMD collective-primitive path so all_reduce/reduce_scatter/all_gather/ alltoall run on MI355X, not just MoRI: - common.sh: cx_build_rccl_tests — clones ROCm/rccl-tests and builds with `make` against /opt/rocm (amdclang++/librccl). It's a nccl-tests fork producing the same <op>_perf binaries and output format, so run_nccl.py parses it unchanged. Validated building + running all 4 ops in-container on MI355X (correctness OK). - run_in_container.sh: run_nccl_suite picks rccl-tests on ROCm (/opt/rocm or hipcc), nccl-tests otherwise; identical op loop + run_nccl.py invocation. - launch_mi355x-amds.sh: honor CX_BENCH (mori default | nccl) instead of forcing mori; same -g N single-node 8-GPU launch. - docs: README/CONTAINERS note the rccl path. B200 already has the nccl path; this makes primitives available on all three SKUs via workflow_dispatch.
…t cancel each other
| if name: | ||
| devices.append(name) | ||
| elif _run(["ibstat", "-l"]): | ||
| devices = [d.strip() for d in _run(["ibstat", "-l"]).splitlines() if d.strip()] |
There was a problem hiding this comment.
ibstat fallback may crash capture
Low Severity
In _rdma, the ibstat -l branch calls _run twice. If the first call succeeds but the second returns None, None.splitlines() raises and env_capture.py aborts before writing provenance JSON for that run.
Reviewed by Cursor Bugbot for commit 2b23573. Configure here.
…on-node launch_gb200-nv.sh now branches on CX_NODES: 1 (default) keeps the single-tray 4-GPU dispatcher path; >1 runs across the NVL72 NVLink fabric (e.g. CX_NODES=2 = 8 GPU) by building nccl-tests MPI=1, running each op across WORLD ranks via `srun --mpi=pmix` (1 GPU/rank) with the MNNVL env, and parsing on the login node — mirroring launch_b200-dgxc-slurm but staying on NVLink instead of IB. Validated on GB200 (2x watchtower-navy trays, 8 GPU): all 4 ops valid, peak busbw all_reduce 822.8 / reduce_scatter 670.6 / all_gather 651.2 / alltoall 625.0 GB/s — ~30% over single-tray and on par with B200 8-GPU NVLink, i.e. MNNVL engaged (not an IB fallback). - common.sh: cx_build_nccl_tests auto-detects MPI_HOME for MPI=1 (Debian OpenMPI headers live under /usr/lib/<arch>/openmpi/include; MPI_HOME=/usr fails). Works x86_64 + aarch64. - launch_b200-dgxc-slurm.sh: fix BUILD_IN_CTR path (.nccl-tests/nccl-tests/build). - workflow: add `nodes` dispatch input -> CX_NODES.
| CX_OPS: ${{ inputs.ops }} | ||
| CX_MIN_BYTES: ${{ inputs.min_bytes }} | ||
| CX_MAX_BYTES: ${{ inputs.max_bytes }} | ||
| CX_NGPUS: ${{ inputs.ngpus }} |
There was a problem hiding this comment.
Workflow ngpus env ignored
Medium Severity
The dispatch job sets CX_NGPUS from the ngpus input, but GB200 and B200 multi-node launchers read CX_GPUS_PER_NODE (defaults 4 and 8) and never CX_NGPUS. Changing ngpus in the workflow does not affect Slurm allocation or world size on those SKUs.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit a3a492c. Configure here.
…ero busbw The first GB200 8-GPU CI run came back green but all-zero busbw: it reused a cached MPI=0 nccl-tests build in the staging dir, and an MPI=0 binary under `srun --mpi=pmix` runs as N standalone world=1 procs (busbw formula -> 0), so every rank printed its own table (232 rows) and check still "passed". - common.sh: cache MPI=0 and MPI=1 builds in separate dirs (nccl-tests vs nccl-tests-mpi) so they never cross-contaminate. - launch_gb200-nv.sh / launch_b200-dgxc-slurm.sh: read the -mpi build dir. - run_nccl.py: a result with peak busbw == 0 is now `invalid` (fails the gate), so a non-communicating run goes red instead of green-zero.
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 2 potential issues.
There are 11 total unresolved issues (including 9 from previous reviews).
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 871086d. Configure here.
| # ---------------------------------------------------------------------------- | ||
| if [ "$NODES" -le 1 ]; then | ||
| # Single tray (4 GPU): generic dispatcher, -g N single process. | ||
| export CX_NGPUS="$GPUS_PER_NODE" |
There was a problem hiding this comment.
Workflow ngpus ignored on GB200
Medium Severity
collectivex-experimental.yml passes CX_NGPUS from the ngpus input, but launch_gb200-nv.sh sizes allocations from CX_GPUS_PER_NODE (default 4) and overwrites CX_NGPUS with that value. Dispatch runs ignore the workflow GPU count unless operators set a different env var manually.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit 871086d. Configure here.
| command -v salloc >/dev/null || cx_die "salloc not found — run on the Slurm login node" | ||
|
|
||
| salloc --partition="$PARTITION" --account="$ACCOUNT" --gres=gpu:"$NGPUS" \ | ||
| --exclusive --time="$TIME_MIN" --no-shell --job-name="$RUNNER_NAME" |
There was a problem hiding this comment.
Single-node B200 ignores nodes
Medium Severity
The dispatch workflow exports CX_NODES, but launch_b200-dgxc.sh always allocates a single node and never reads CX_NODES. Choosing b200-dgxc with nodes greater than one still runs an 8-GPU single-node job, not the multinode Slurm adapter.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit 871086d. Configure here.
1 participant
Adds CollectiveX under
experimental/CollectiveX/— a cross-vendor collective / expert-parallel benchmark — plus an orchestration-only workflow.What it adds
launchers/launch_<sku>.sh, thelaunch_${RUNNER_NAME%%_*}.shconvention) that run any benchmark via aCX_BENCHselector (nccl|deepep|all) through a sharedlaunchers/run_in_container.sh.run_nccl.py(stock nccl-tests → parsed flat JSON),run_deepep.py(DeepEP dispatch/combine, normal mode),env_capture.py(Layer-0 provenance),plot.py. Every result is correctness-gated and carries a topology-awarecomparison_key.lmsysorg/sglang@sha256:4219…, amd64+arm64); DeepEP viarebuild-deepep. SeeCONTAINERS.md..github/workflows/collectivex-experimental.yml—pushtocollectivex(pathsexperimental/CollectiveX/**) → GB200 NCCL smoke;workflow_dispatch→ chosensku+benchmark(B200, DeepEP, larger sweeps). Logic stays underexperimental/.Validated on hardware
bash -n,py_compile, actionlint, parser fixtures.Notes / deferred
CONTAINERS.md.v0.5.11-cu130).plan.md, not built.Note
Low Risk
Changes are isolated to
experimental/CollectiveX/and a read-only workflow; no production benchmark matrix or serving launchers are modified. Risk is mainly operational (self-hosted GPU time, Slurm/enroot failures) rather than app or security impact.Overview
Introduces CollectiveX under
experimental/CollectiveX/— an experimental cross-vendor collective and MoE EP benchmark — plus orchestration-only.github/workflows/collectivex-experimental.yml. Production serving paths are untouched.Benchmark stack:
run_nccl.pywraps nccl-tests/rccl-tests into provenance-tagged JSON;run_deepep.pyandrun_mori.pyadd correctness-gated DeepEP and AMD MoRI dispatch/combine;env_capture.py,summarize.py, andplot.pyhandle environment capture, CI summaries, and plots. Results use topology-awarecomparison_keys so unlike fabrics are not merged blindly.Execution: Per-SKU Slurm launchers (
launch_b200-dgxc.sh,launch_gb200-nv.sh,launch_b200-dgxc-slurm.sh,launch_mi355x-amds.sh) follow the samelaunch_${RUNNER_NAME%%_*}.shpattern as serving, with sharedcommon.sh(enroot squash by tag, optionalCX_STAGE_DIRrsync, in-container nccl/rccl builds).CX_BENCHselectsnccl,deepep,mori, orallviarun_in_container.sh.CI: Push to
collectivexruns MI355X MoRI onmi355xrunners;workflow_dispatchpicks SKU and benchmark (GB200/B200 NCCL, DeepEP, etc.), writes markdown to the job summary, and uploads gitignoredresults/*.jsonas artifacts.Reviewed by Cursor Bugbot for commit 871086d. Bugbot is set up for automated code reviews on this repo. Configure here.