[https://nvbugs/5452463][doc] update disagg doc about UCX_MAX_RNDV_RAILS #7205
Conversation
Signed-off-by: zhengd-nv <200704041+zhengd-nv@users.noreply.github.com>
📝 WalkthroughWalkthroughDocumentation update to disaggregated-service: adds the Changes
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~3 minutes Possibly related PRs
Suggested reviewers
Tip 🔌 Remote MCP (Model Context Protocol) integration is now available!Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats. 📜 Recent review detailsConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro 💡 Knowledge Base configuration:
You can enable these sources in your CodeRabbit configuration. 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
✨ Finishing Touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (4)
docs/source/advanced/disaggregated-service.md (4)
39-41: Define or link the “TEP” acronym on first mentionReaders unfamiliar with TEP may not know what it stands for. Either expand it on first mention or link to the section that defines/enables it.
If “TEP” is already defined elsewhere in the docs, please point this section to it with a link to improve discoverability.
39-41: Operator guidance: add “when to use 1 vs 2” rule-of-thumbA short decision aid helps users:
- Use
2when each GPU has sufficient dedicated NIC bandwidth (e.g., no cross-GPU NIC sharing).- Use
1when context+generation both enable TEP and concurrent TP KV transfers cause NIC contention across GPUs.Optionally suggest how to inspect topology and NIC/GPU mapping (run on the hosts):
nvidia-smi topo -mto see GPU↔NIC proximityibdev2netdev(OFED) to map HCAs to net devices
80-80: Polish the FAQ question phrasingMinor readability tweak.
-*Q. When my servers are running on different NVLink domains, some servers hang or have a lower performance. How to fix that?* +*Q. When servers are in different NVLink domains, some servers hang or performance drops. How can I address this?*
39-41: Confirm UCX_MAX_RNDV_RAILS default and add referenceUCX_MAX_RNDV_RAILS controls how many network “rails” (NIC devices) UCX will use for rendezvous (large-message) transfers. Across UCX releases, the default is 2 rails (ucx-shamisp.readthedocs.io, docs.nvidia.com). Please add a reference to the official UCX documentation (e.g. the OpenUCX FAQ or NVIDIA HPC-X docs) here, or note that TRT-LLM recommends using the default of 2 unless overridden.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
docs/source/advanced/disaggregated-service.md(2 hunks)
🧰 Additional context used
🪛 LanguageTool
docs/source/advanced/disaggregated-service.md
[grammar] ~39-~39: There might be a mistake here.
Context: ...try to use 2 ib NIC device for each GPU. If context and geneneration instances ar...
(QB_NEW_EN)
[grammar] ~40-~40: Ensure spelling is correct
Context: ...NIC device for each GPU. If context and geneneration instances are both TEP, the KV cache tr...
(QB_NEW_EN_ORTHOGRAPHY_ERROR_IDS_1)
[grammar] ~40-~40: There might be a mistake here.
Context: ...t to 1 helps to improve efficiency in such case. ## Troubleshooting and FAQ ### ...
(QB_NEW_EN)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (1)
docs/source/advanced/disaggregated-service.md (1)
39-41: Good addition: actionable knob for NIC contention under TEP + TP concurrencyAdding
UCX_MAX_RNDV_RAILShere is helpful and gives operators a concrete mitigation when context and generation both enable TEP and TP ranks contend for NICs. Nice catch to surface this in the Environment Variables section.
|
/bot skip --comment "only update doc" |
|
PR_Github #16500 [ skip ] triggered by Bot |
|
PR_Github #16500 [ skip ] completed with state |
4 participants
Summary by CodeRabbit
Description
Test Coverage
In disaggregated serving, when both ctx and gen enabled TEP, some NIC device may be occupied by multiple GPUs due to parallel cache transfer, and lead to slow cache transmission. The env can solve the problem if the user encountered such a case.
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.