Fix errors and update prefill delayer

[fzyzcjy]

Motivation

cc @merrymercy

prev: #14201

e.g. see mem leak on some cases

Modifications

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.
• Follow the SGLang code style guidance.

Review Process

1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments (/tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci) or contact authorized users to do so.
4. After green CI and required approvals, ask Merge Oncalls to merge.

[gemini-code-assist]

Summary of Changes

Hello @fzyzcjy, 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 significantly refactors and enhances the prefill delay mechanism within the SGLang scheduler. The core component responsible for managing prefill delays has been renamed and its internal logic revised to improve clarity and potentially efficiency in distributed environments. These changes are seamlessly integrated into the existing scheduler, and new comprehensive tests are introduced to validate the impact on system throughput.

Highlights

• Refactored Prefill Delay Logic: The SchedulerEnhancer class has been renamed to PrefillDelayer and its internal logic for deciding when to delay prefill requests has been updated and clarified, now focusing on the waiting queue length across distributed processes.
• Scheduler Integration Update: The main scheduler (scheduler.py) has been updated to use the new PrefillDelayer class and its should_allow_prefill method, streamlining the prefill decision-making process.
• New Throughput Tests: A dedicated test file (test_prefill_delayer.py) has been added to benchmark the throughput performance with and without the prefill delayer enabled, specifically for distributed attention setups.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 refactors the SchedulerEnhancer into PrefillDelayer, improving clarity and fixing a bug in its initialization. The logic for delaying prefill operations is updated to be based on the waiting queue length, which is a more direct measure of workload imbalance. A new test suite is added to validate the functionality and performance impact of this feature. I've identified a couple of areas for improvement, including a potential resource leak in the new test file and some stylistic enhancements in the boolean comparisons.