[Cherry-pick v1.12] add hcclrank job plugin#4555
Conversation
Signed-off-by: wangdongyang1 <wangdongyang1@sensetime.com>
There was a problem hiding this comment.
Summary of Changes
Hello @kingeasternsun, 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 a new hcclrank job plugin to the Volcano project, primarily to enable and streamline distributed training on Ascend NPU SuperPod. The plugin automatically assigns appropriate HCCL ranks to pods within a job, either by leveraging existing environment variables or by calculating ranks based on task indices and replica counts. This enhancement aims to simplify the deployment and management of distributed AI training jobs on specialized hardware.
Highlights
- New HCCLRank Job Plugin: A new
hcclrankjob plugin has been introduced to support distributed training workloads, specifically targeting Ascend NPU SuperPod environments. This plugin is responsible for calculating and assigning a uniquehccl/rankIndexto pods during their creation. - Enhanced Job Helper Functions: Two new helper functions,
GetTaskIndexOfPodandGetTaskReplicasUnderJob, have been added topkg/controllers/job/helpers/helpers.go. These functions provide standardized ways for plugins to retrieve crucial information about pod task indices and task replica counts within a job, facilitating accurate rank calculation. - Plugin Integration and Registration: The newly developed
hcclrankplugin has been integrated into the Volcano job plugin factory. This registration ensures that the plugin is discoverable and can be activated for jobs requiring HCCL rank assignment.
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 introduces the hcclrank job plugin to support Ascend NPU SuperPod by automatically assigning rank annotations to pods. It also adds two new helper functions, GetTaskIndexOfPod and GetTaskReplicasUnderJob, with corresponding unit tests. The overall implementation is solid and well-tested.
My main feedback is to add a check in the hcclrank plugin to ensure the master task is defined in the job before calculating ranks for worker pods. This will make the plugin more robust against misconfigurations. Please see the detailed comment.
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: hwdef The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/lgtm |
What type of PR is this?
Add the hcclrank job plugin
What this PR does / why we need it:
To solve #4523
Which issue(s) this PR fixes:
Fixes #4523
Special notes for your reviewer:
Does this PR introduce a user-facing change?