EPLB: prefer to use physical experts in the same node

[acelyc111]

Motivation

Because of NVLink, intranode communication is faster than that of internode, when it's possible to use physical experts on the same node, we'd prefer to route to these GPU ranks.

Modifications

When setting --ep-dispatch-algorithm static, select the experts on the same GPU at first if possible, then select experts on the same node if possible, at last, select the experts randomly as a fall back method.

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.

[gemini-code-assist]

Summary of Changes

Hello @acelyc111, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refines the expert dispatching mechanism within the system. It introduces a hierarchical preference for expert selection, aiming to improve performance by minimizing cross-host communication. The changes ensure that the system first attempts to utilize experts on the same GPU, then on the same physical host, before considering other available experts.

Highlights

• Enhanced Expert Locality Preference: The expert dispatching logic now prioritizes physical experts on the same GPU, then on the same host, before resorting to fair random choices.
• Introduction of Host-Level Expert Tracking: New variables and a helper function (_compute_host_id_of_physical_expert) have been added to correctly identify and manage physical experts at the host level.
• Refinement of GPU Expert Calculation: The calculation for GPU-local physical experts has been clarified and its corresponding helper function parameter updated for better specificity.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a valuable optimization by prioritizing physical experts on the same host, after checking for same-GPU experts. This should improve performance by leveraging locality. The implementation logic is sound. However, a key issue is the hardcoding of the number of GPUs per host to 8. This limits the feature's applicability and should be made dynamic. My review comments provide specific suggestions to resolve this.

[acelyc111]

@fzyzcjy Please take a look, thanks!