[IB-1920][doc] Update Perf_Overview.md with Benchmarking Results for Release 1.1 #9723
Conversation
|
/bot skip docs |
GitHub Bot Help
Provide a user friendly way for developers to interact with a Jenkins server. Run See details below for each supported subcommand. Details
Launch build/test pipelines. All previously running jobs will be killed.
kill
Kill all running builds associated with pull request. skip
Skip testing for latest commit on pull request. reuse-pipeline
Reuse 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. |
📝 WalkthroughWalkthroughDocumentation reorganization of performance benchmark content with updated model listings, new section structure with Table of Contents and per-GPU configurations, revised benchmark tables with updated values, and clarified dataset generation and workflow descriptions. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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. Comment |
![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
🧹 Nitpick comments (4)
docs/source/developer-guide/perf-overview.md (4)
10-10: Consider adding clarity about which configurations were tested.Line 10 notes that "Not all configurations were tested for all GPUs," which is helpful. However, consider clarifying whether users can infer from the tables which combinations were tested or if they should consult another resource for specifics.
87-95: Benchmark tables lack clarity on units and precision.The performance tables show throughput numbers (e.g., "25,926", "53,761") but the units (output tokens per second per GPU) are mentioned in a separate section. Consider adding a subtle footnote or column header to each table or repeating the unit notation near the first data row to reduce user confusion, especially for users who scan tables out of sequence.
Also applies to: 102-110, 117-124, 147-155, 177-185, 209-217
241-243: Emphasize model list limitations more visibly in the procedure.The note states "Only the models shown in the table above are supported by this workflow." However, it uses a generic
{note}block that may not stand out to users who proceed directly to the "Command Overview" section. Consider using a warning or highlight to ensure users don't attempt unsupported models.
249-254: Add reference to PyTorch backend documentation for clarity.Line 249 states "Testing was performed using the PyTorch backend - this workflow does not require an engine to be built." Consider adding a link or reference to the PyTorch backend documentation to help users understand when to use this approach versus building a TensorRT engine.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/source/developer-guide/perf-overview.md(4 hunks)
🧰 Additional context used
🧠 Learnings (10)
📓 Common learnings
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.
Learnt from: farshadghodsian
Repo: NVIDIA/TensorRT-LLM PR: 7101
File: docs/source/blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.md:36-36
Timestamp: 2025-08-21T00:16:56.457Z
Learning: TensorRT-LLM container release tags in documentation should only reference published NGC container images. The README badge version may be ahead of the actual published container versions.
📚 Learning: 2025-08-26T09:49:04.956Z
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7192
File: tests/integration/test_lists/test-db/l0_dgx_b200.yml:56-72
Timestamp: 2025-08-26T09:49:04.956Z
Learning: In TensorRT-LLM test configuration files, the test scheduling system handles wildcard matching with special rules that prevent duplicate test execution even when the same tests appear in multiple yaml files with overlapping GPU wildcards (e.g., "*b200*" and "*gb200*").
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-09-09T09:40:45.658Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 7645
File: tests/integration/test_lists/qa/llm_function_core.txt:648-648
Timestamp: 2025-09-09T09:40:45.658Z
Learning: In TensorRT-LLM test lists, it's common and intentional for the same test to appear in multiple test list files when they serve different purposes (e.g., llm_function_core.txt for comprehensive core functionality testing and llm_function_core_sanity.txt for quick sanity checks). This duplication allows tests to be run in different testing contexts.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-08-11T20:09:24.389Z
Learnt from: achartier
Repo: NVIDIA/TensorRT-LLM PR: 6763
File: tests/integration/defs/triton_server/conftest.py:16-22
Timestamp: 2025-08-11T20:09:24.389Z
Learning: In the TensorRT-LLM test infrastructure, the team prefers simple, direct solutions (like hard-coding directory traversal counts) over more complex but robust approaches when dealing with stable directory structures. They accept the maintenance cost of updating tests if the layout changes.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
Repo: NVIDIA/TensorRT-LLM PR: 6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-11-27T09:23:18.742Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 9511
File: tests/integration/defs/examples/serve/test_serve.py:136-186
Timestamp: 2025-11-27T09:23:18.742Z
Learning: In TensorRT-LLM testing, when adding test cases based on RCCA commands, the command format should be copied exactly as it appears in the RCCA case, even if it differs from existing tests. For example, some RCCA commands for trtllm-serve may omit the "serve" subcommand while others include it.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
Repo: NVIDIA/TensorRT-LLM PR: 6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM's bench configuration, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which is a Dict[str, Any] that can contain default values including `cuda_graph_config`, making the fallback `llm_args["cuda_graph_config"]` safe to use.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-08-21T00:16:56.457Z
Learnt from: farshadghodsian
Repo: NVIDIA/TensorRT-LLM PR: 7101
File: docs/source/blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.md:36-36
Timestamp: 2025-08-21T00:16:56.457Z
Learning: TensorRT-LLM container release tags in documentation should only reference published NGC container images. The README badge version may be ahead of the actual published container versions.
Applied to files:
docs/source/developer-guide/perf-overview.md
📚 Learning: 2025-08-13T11:07:11.772Z
Learnt from: Funatiq
Repo: NVIDIA/TensorRT-LLM PR: 6754
File: tests/integration/test_lists/test-db/l0_a30.yml:41-47
Timestamp: 2025-08-13T11:07:11.772Z
Learning: In TensorRT-LLM test configuration files like tests/integration/test_lists/test-db/l0_a30.yml, TIMEOUT values are specified in minutes, not seconds.
Applied to files:
docs/source/developer-guide/perf-overview.md
⏰ 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 (6)
docs/source/developer-guide/perf-overview.md (6)
62-235: Well-structured performance summary with clear organization.The reorganization with a Table of Contents (lines 68–80), per-model sections with anchor links, and per-GPU-configuration RTX tables greatly improves readability. The new structure makes it easy for users to navigate to the benchmark results they need.
129-140: RTX configuration tables introduce additional clarity for multi-GPU setups.The new RTX-specific sections (LLaMA, Qwen3) with TP/PP configurations provide valuable guidance for users deploying on RTX hardware with multiple GPUs. The note about "best configuration per GPU count based on throughput per GPU" is helpful.
Also applies to: 160-170, 191-202, 222-235
276-278: Dataset preparation command is comprehensive and well-documented.The command for generating synthetic datasets is clear with inline parameter descriptions. Users should easily understand the purpose of each flag and adjust values as needed.
292-299: Dataset request counts table improves reproducibility.The table specifying request counts for each ISL/OSL combination directly addresses the need for reproducible benchmarks by justifying why different request volumes are used. This is excellent documentation practice.
307-333: CUDA graph configuration options are comprehensive.The example LLM options YAML files (for both dense and MoE models) provide clear guidance on tuning batch sizes and padding for CUDA graphs. The distinction between dense and MoE configurations is helpful.
364-365: KV cache tuning note adds practical guidance.Line 364–365 notes that using
--kv_cache_free_gpu_mem_fraction 0.95can improve performance but suggests fallback to 0.90 or lower if OOM is encountered. This practical guidance helps users reproduce the benchmark results without running into OOM issues unexpectedly.
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
zbpatel
force-pushed
the
user/zpatel/perf_overview_1.1
branch
from
be74f33 to
fb3d690
Compare
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
zbpatel
force-pushed
the
user/zpatel/perf_overview_1.1
branch
2 times, most recently
from
f339c73 to
d625baf
Compare
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
zbpatel
force-pushed
the
user/zpatel/perf_overview_1.1
branch
from
d625baf to
87c08e4
Compare
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
zbpatel
changed the title
|
I did some cleaning and went in and updated the configs for what we used in the 1.1 testing. I think this document is ready for review. @kaiyux do you mind taking a look at this? |
|
I'll be online for the next 2 hours or so, if anyone wants to make changes after I log off, feel free to commit directly to the branch so that we don't block the release. |
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
chzblych
added
the
Release Blocker
Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
…ated with links to code Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
|
/bot skip docs |
GitHub Bot Help
Provide a user friendly way for developers to interact with a Jenkins server. Run See details below for each supported subcommand. Details
Launch build/test pipelines. All previously running jobs will be killed.
kill
Kill all running builds associated with pull request. skip
Skip testing for latest commit on pull request. reuse-pipeline
Reuse 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. |
|
/bot skip --comment docs |
|
PR_Github #27563 [ skip ] triggered by Bot. Commit: |
|
PR_Github #27563 [ skip ] completed with state |
|
/bot skip --comment "docs / retrying" |
|
PR_Github #27566 [ skip ] triggered by Bot. Commit: |
|
PR_Github #27566 [ skip ] completed with state |
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…Release 1.1 (NVIDIA#9723) Signed-off-by: Zachary Patel <22306219+zbpatel@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
6 participants
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.
Description
This MR updates the performance documentation with numbers and updated benchmark instructions from the release 1.1 performance benchmarks.
Notable Changes:
Test Coverage
None / Docs
PR Checklist
Please review the following before submitting your PR:
PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
Update tava architecture diagram if there is a significant design change in PR.
The reviewers assigned automatically/manually are appropriate for the PR.
[ x] Please check this after reviewing the above items as appropriate for this PR.
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.
Details
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.