doc: Add README for wide EP #6356
Conversation
📝 Walkthrough## Walkthrough
The changes update documentation and configuration related to Mixture-of-Experts (MoE) and Wide Expert Parallelism (Wide-EP) in TensorRT-LLM. Documentation files are corrected and expanded, and the MoE configuration structure in a YAML generator script is refactored for clarity by nesting related settings under a dedicated dictionary. Additionally, SLURM submission scripts were modified to print reminders at runtime instead of using comments, with some job submission commands removed or commented out.
## Changes
| Files/Group | Change Summary |
|-----------------------------------------------------|---------------------------------------------------------------------------------------------------------------|
| **Documentation fixes and additions** <br> `docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md` | Fixed two broken anchor links in Table of Contents and added a new subsection header for clarity. |
| **MoE configuration refactor** <br> `examples/disaggregated/slurm/gen_yaml.py` | Refactored MoE backend/load balancer configuration to nest under a new `moe_config` dictionary; grouped kv_cache settings under `kv_cache_config`. |
| **Wide Expert Parallelism documentation** <br> `examples/wide_ep/README.md` | Added new README introducing Wide-EP, its challenges, solutions, configuration, and quick start instructions. |
| **SLURM submission script updates** <br> `examples/disaggregated/slurm/submit.sh` | Replaced initial comment with echo statement printing SLURM parameter reminder at runtime. |
| **Wide-EP SLURM job submission script changes** <br> `examples/wide_ep/slurm_scripts/submit.sh` | Replaced comment block with echo statements printing configuration instructions at runtime; removed a job submission loop and commented out some sbatch commands. |
## Estimated code review effort
🎯 2 (Simple) | ⏱️ ~8 minutesSuggested labels
Suggested reviewers
Note ⚡️ Unit Test Generation is now available in beta!Learn more here, or try it out under "Finishing Touches" below. 📜 Recent review detailsConfiguration used: .coderabbit.yaml 📒 Files selected for processing (1)
✅ Files skipped from review due to trivial changes (1)
✨ Finishing Touches🧪 Generate unit tests
🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
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: 2
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md(3 hunks)examples/disaggregated/slurm/gen_yaml.py(2 hunks)examples/wide_ep/README.md(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, even if only one class or function from a module is used.
Python filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL).
Python constants should use upper snake_case (e.g., MY_CONSTANT).
Avoid shadowing variables declared in an outer scope in Python.
Initialize all externally visible members of a Python class in the constructor.
For interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for Python classes and functions, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
examples/disaggregated/slurm/gen_yaml.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
examples/disaggregated/slurm/gen_yaml.py
🔇 Additional comments (8)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md (3)
6-6: LGTM: Anchor link correctionThe missing hyphen in the anchor link has been properly added to match the section header format.
18-19: LGTM: Anchor link correctionsBoth anchor links have been corrected with missing hyphens to properly reference their respective subsections.
519-521: LGTM: Document structure improvementThe new subsection header properly organizes the content under "Reproducing steps" and matches the existing formatting style.
examples/disaggregated/slurm/gen_yaml.py (2)
205-207: LGTM: Configuration structure improvementThe restructuring of MoE settings under a dedicated
moe_configdictionary improves organization and aligns with the Wide-EP framework. The nested structure is cleaner than the previous flat configuration approach.
247-248: LGTM: Consistent configuration nestingThe load balancer configuration is now properly nested under
moe_config, maintaining consistency with the overall MoE configuration restructuring. The key simplification frommoe_load_balancertoload_balanceris appropriate within the MoE context.examples/wide_ep/README.md (3)
1-19: LGTM: Comprehensive introductionThe introduction effectively explains the Wide-EP feature, clearly articulates the challenges it addresses, and outlines the solutions provided. The content is well-structured and informative.
38-50: LGTM: Clear load balancer documentationThe load balancer configuration is well-documented with helpful parameter descriptions and a clear distinction between offline and online modes. The YAML example is consistent with the expected configuration structure.
65-67: Valid subdirectory referencesVerified that both referenced paths exist under
examples/wide_ep/:
- examples/wide_ep/ep_load_balancer/
- examples/wide_ep/slurm_scripts/
No updates required in the README.
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
kaiyux
force-pushed
the
user/kaiyu/wideep_readme
branch
from
3ab8f96 to
67fb3de
Compare
coderabbitai
bot
added
Community want to contribute
Signed-off-by: Kaiyu Xie <[email protected]>
coderabbitai
bot
added
the
Community want to contribute
|
/bot skip --comment "doc changes" |
|
PR_Github #13289 [ skip ] triggered by Bot |
|
PR_Github #13289 [ skip ] completed with state |
Signed-off-by: Kaiyu Xie <[email protected]> Signed-off-by: Lanyu Liao <[email protected]>
4 participants
Summary by CodeRabbit
Documentation
Refactor
Chores
Description
Preview: https://github.com/kaiyux/TensorRT-LLM/tree/user/kaiyu/wideep_readme/examples/wide_ep
Test Coverage
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.