Optimize queueAllocatable calculation by excluding allocated resources from unschedulable nodes #4562
Conversation
There was a problem hiding this comment.
Summary of Changes
Hello @dafu-wu, 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 introduces an optimization to the queueAllocatable calculation within the scheduler's capacity plugin. The primary goal is to improve the accuracy of queue-level resource accounting by excluding resources allocated on nodes that are unschedulable or not ready. This prevents the scheduler from making decisions based on unusable capacity, leading to more precise resource management and fairer scheduling. The change ensures that only truly available resources contribute to a queue's allocatable capacity, without impacting already scheduled workloads.
Highlights
- Refined Resource Accounting: A new field,
allocatedWithoutSchedulingDisabledNode, has been introduced in thequeueAttrstruct. This field specifically tracks resources allocated on nodes that are currently schedulable and ready, providing a more accurate view of available capacity. - Optimized Scheduling Logic: The core
queueAllocatablefunction, which determines if a queue can accept new tasks, now utilizes theallocatedWithoutSchedulingDisabledNodemetric. This change ensures that scheduling decisions are based only on resources from healthy and available nodes, preventing over-commitment due to resources on unschedulable or unready nodes. - Enhanced Node State Checks: New helper functions,
isNodeUnschedulableandisNodeNotReady, have been added. These functions are used during the resource aggregation process to precisely filter out resources from nodes that are marked as unschedulable or are not in a ready state, ensuring the accuracy of the new resource metric.
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
-
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. ↩
There was a problem hiding this comment.
Code Review
This pull request correctly identifies and fixes an issue where resources on unschedulable nodes were counted towards a queue's allocation, leading to inaccurate scheduling. The approach of introducing a new field allocatedWithoutSchedulingDisabledNode is sound. However, the implementation has a critical flaw that could lead to a panic due to a nil pointer, and there are opportunities to improve maintainability by reducing code duplication. I've provided detailed comments on these points.
6509e91 to
852f83e
Compare
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
852f83e to
bf509d7
Compare
…urces from unschedulable nodes Signed-off-by: dafu <wuchengyi2006@163.com>
c8ee27f to
fe45ad9
Compare
|
@JesseStutler @Monokaix @shinytang6 @alcorj-mizar can you help to review this PR? |
JesseStutler
left a comment
There was a problem hiding this comment.
I think we may also have same problems in proportion plugin
@JesseStutler can you help to modify it? I didn't see any relevant code |
…Node Signed-off-by: dafu <wuchengyi2006@163.com>
655c925 to
445a1fe
Compare
|
@JesseStutler @caniszczyk can you merge this pr or fix this issue asap? It's block us🙏 |
|
We can revert this one #4373 |
|
@dafu-wu: PR needs rebase. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
Resolved #4575 |
What type of PR is this?
/kind bug
/kind cleanup
What this PR does / why we need it:
This PR optimizes the calculation of
queueAllocatableby excluding allocated resources that come from nodes marked as unschedulable.Without this fix, queue-level resource accounting may include unusable allocations, leading to inaccurate scheduling decisions and potential over-commit.
Which issue(s) this PR fixes:
Fixes #
Special notes for your reviewer:
Does this PR introduce a user-facing change?