[None][doc] Display tech blog for nvidia.github.io domain. #7241
Conversation
📝 WalkthroughWalkthroughUpdated three tech blog Markdown files to replace relative image paths with absolute GitHub raw URLs. Modified docs index to switch the Blogs toctree from :hidden: to :glob: and to include all tech blog files via a wildcard instead of explicit entries. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested labels
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. ✨ 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
|
nv-guomingz
changed the title
Signed-off-by: nv-guomingz <[email protected]>
nv-guomingz
force-pushed
the
user/guomingz/display_all_tech_blogs
branch
from
1a23dcd to
c99ac2c
Compare
|
/bot run |
nv-guomingz
changed the title
|
PR_Github #16521 [ run ] triggered by Bot |
![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
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (6)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md (6)
235-235: Incorrect figure reference (Figure 11 → Figure 12).The text points to Figure 11, but the high-level design diagram is Figure 12.
- * EP communication kernels. In Figure 11, those are the **Dispatch** and **Combine** components. + * EP communication kernels. In Figure 12, those are the **Dispatch** and **Combine** components.
231-231: Fix “Placemen t” typo in bold label.Stray emphasis breaks the word.
- * Orchestrate the process (**Update Weights \& Placemen**t component) to update and reload the MoE weights... + * Orchestrate the process (**Update Weights \& Placement** component) to update and reload the MoE weights...-For the **Update Weights \& Placemen**t component, we identified two design choices: +For the **Update Weights \& Placement** component, we identified two design choices:Also applies to: 241-241
324-324: Grammar: “take of advantage” → “take advantage of”.-We also wanted those kernels, for GB200, to take of advantage of the MNNVL's memory bandwidth. +We also wanted those kernels, for GB200, to take advantage of the MNNVL's memory bandwidth.
293-293: Pluralization: “all the method” → “all the methods”.- * Need 93 Expert Update / Iter, all the method can satisfy the latency goal. + * Need 93 Expert Update / Iter, all the methods can satisfy the latency goal.
399-399: Duplicate word: “how how” → “how”.-... learn more details as well as how how to run through the Offline EP Load Balancer in E2E approach. +... learn more details as well as how to run through the Offline EP Load Balancer in an E2E approach.
1-718: Critical: Fix media references, accessibility attributes, layout constraints, and typos before mergePlease address the following issues uncovered by the verification script:
Relative media reference remains
- docs/source/blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.md:66 uses
—update to the absolute raw GitHub URL.Missing
altattributes on<img>tags
Several blog files (e.g., blog8, blog7, blog4, blog5, blog1, blog2, advanced/expert-parallelism.md) contain<img src="https://github.com/%E2%80%A6">without anyalt=. Ensure every<img>has a meaningfulaltdescription.Hard-coded
widthattributes on images
Instances in blog7, blog8, blog4, blog5, etc., usewidth="auto",width="80%",width="500", etc. Remove or replace these with responsive styling (CSS or markdown syntax) to avoid layout issues.Known typos still present in blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md
- Lines ~230 & 241: “Placemen t” → “Placement”
- Line ~323: “take of advantage” → “take advantage of”
- Line ~399: duplicate “how how” → single “how”
After fixing, rerun the checks to confirm:
- No
../media/patterns in anydocs/.- No
<img>withoutalt.- No hard-coded
widthattributes.- All typos corrected.
🧹 Nitpick comments (4)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md (4)
261-261: Avoid fixed pixel widths; make images responsive.Hard-coded pixel widths are not mobile-friendly. Let CSS handle sizing or use max-width: 100%.
For a minimal fix, remove width attributes:
- <img src="https://github.com/.../tech_blog4_Picture14.png" width="500" > + <img alt="Figure 14: high-level topology of GB200 system" src="https://github.com/.../tech_blog4_Picture14.png">Or, if you must constrain width, prefer a style that preserves responsiveness:
- <img src="https://github.com/.../tech_blog4_Picture14.png" width="500" > + <img alt="Figure 14: high-level topology of GB200 system" src="https://github.com/.../tech_blog4_Picture14.png" style="max-width:100%;height:auto;">Also applies to: 273-273, 501-501, 511-511
85-85: Consider pinning image URLs or switching to Sphinx-managed relative assets to avoid future breakage.External links to “.../raw/main/...” may 302-redirect and can break if files move. Two safer options:
- Preferred: convert to MyST/Sphinx figures with relative paths so Sphinx copies and rewrites URLs.
- Alternative: pin to a specific commit using raw.githubusercontent.com to avoid redirects and drift.
Examples:
A) MyST figure (Sphinx-managed, relative asset):
```{figure} ../media/tech_blog4_Picture1.png :align: center :name: fig:ep-figure-1 Figure 1: The routed token count from rank 0 to all the ranks (including rank 0), for decode iteration 1950, and all the MoE layersB) Pin to a commit (replace <COMMIT_SHA> with the image commit): ```diff - <img alt="..." src="https://github.com/NVIDIA/TensorRT-LLM/raw/main/docs/source/blogs/media/tech_blog4_Picture1.png"> + <img alt="..." src="https://raw.githubusercontent.com/NVIDIA/TensorRT-LLM/<COMMIT_SHA>/docs/source/blogs/media/tech_blog4_Picture1.png">Also applies to: 96-96, 105-105, 114-114, 124-124, 142-142, 151-151, 161-161, 170-170, 179-179, 188-188, 221-221, 250-250, 261-261, 273-273, 410-410, 417-417, 428-428, 435-435, 444-444, 451-451, 476-476, 487-487, 501-501, 511-511
505-506: Minor consistency nit: “Group GEMM” vs “GroupGEMM”.Use one style consistently throughout the doc (both variants appear).
Example:
- In Figure 24, it can be observed that by increasing the EP size from 4 to 72, the MoE Group GEMM computation time gets reduced, while the EP communication time ... + In Figure 24, it can be observed that by increasing the EP size from 4 to 72, the MoE GroupGEMM computation time gets reduced, while the EP communication time ...
85-85: Optional: drop deprecated align attributes and rely on CSS or figcaption for centering.is deprecated HTML. You can center via CSS or figure styling. If you adopt figcaption (see earlier comment), remove the extra
.
No diff provided here since it’s coupled to the figcaption change above—happy to generate a complete conversion if you choose that route.
Also applies to: 96-96, 105-105, 114-114, 124-124, 142-142, 151-151, 161-161, 170-170, 179-179, 188-188, 221-221, 250-250, 261-261, 273-273, 410-410, 417-417, 428-428, 435-435, 444-444, 451-451, 476-476, 487-487, 501-501, 511-511
📜 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 (4)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md(22 hunks)docs/source/blogs/tech_blog/blog7_NGram_performance_Analysis_And_Auto_Enablement.md(7 hunks)docs/source/blogs/tech_blog/blog8_Scaling_Expert_Parallelism_in_TensorRT-LLM_part2.md(6 hunks)docs/source/index.rst(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (3)
- docs/source/blogs/tech_blog/blog7_NGram_performance_Analysis_And_Auto_Enablement.md
- docs/source/index.rst
- docs/source/blogs/tech_blog/blog8_Scaling_Expert_Parallelism_in_TensorRT-LLM_part2.md
🧰 Additional context used
🧠 Learnings (1)
📚 Learning: 2025-08-20T07:43:36.447Z
Learnt from: ChristinaZ
PR: NVIDIA/TensorRT-LLM#7068
File: cpp/tensorrt_llm/kernels/moeTopKFuncs.cuh:169-172
Timestamp: 2025-08-20T07:43:36.447Z
Learning: In TensorRT-LLM MOE kernels, when processing up to 128 experts across 32 threads, each thread handles at most 4 experts (N < 5 constraint), where N represents candidates per thread rather than total system capacity.
Applied to files:
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md
🪛 markdownlint-cli2 (0.17.2)
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md
85-85: Images should have alternate text (alt text)
(MD045, no-alt-text)
96-96: Images should have alternate text (alt text)
(MD045, no-alt-text)
105-105: Images should have alternate text (alt text)
(MD045, no-alt-text)
114-114: Images should have alternate text (alt text)
(MD045, no-alt-text)
124-124: Images should have alternate text (alt text)
(MD045, no-alt-text)
142-142: Images should have alternate text (alt text)
(MD045, no-alt-text)
151-151: Images should have alternate text (alt text)
(MD045, no-alt-text)
161-161: Images should have alternate text (alt text)
(MD045, no-alt-text)
170-170: Images should have alternate text (alt text)
(MD045, no-alt-text)
179-179: Images should have alternate text (alt text)
(MD045, no-alt-text)
188-188: Images should have alternate text (alt text)
(MD045, no-alt-text)
221-221: Images should have alternate text (alt text)
(MD045, no-alt-text)
250-250: Images should have alternate text (alt text)
(MD045, no-alt-text)
261-261: Images should have alternate text (alt text)
(MD045, no-alt-text)
273-273: Images should have alternate text (alt text)
(MD045, no-alt-text)
410-410: Images should have alternate text (alt text)
(MD045, no-alt-text)
417-417: Images should have alternate text (alt text)
(MD045, no-alt-text)
428-428: Images should have alternate text (alt text)
(MD045, no-alt-text)
435-435: Images should have alternate text (alt text)
(MD045, no-alt-text)
444-444: Images should have alternate text (alt text)
(MD045, no-alt-text)
451-451: Images should have alternate text (alt text)
(MD045, no-alt-text)
476-476: Images should have alternate text (alt text)
(MD045, no-alt-text)
487-487: Images should have alternate text (alt text)
(MD045, no-alt-text)
501-501: Images should have alternate text (alt text)
(MD045, no-alt-text)
511-511: Images should have alternate text (alt text)
(MD045, no-alt-text)
⏰ 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/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md (1)
85-85: Switch to absolute image URLs achieves the stated goal (images render on nvidia.github.io).Using absolute GitHub-hosted URLs should fix the broken images on the published site. Good call for an immediate doc-site fix.
docs/source/blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.md
Show resolved
Hide resolved
|
PR_Github #16521 [ run ] completed with state |
3 participants
Summary by CodeRabbit
Description
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.