[Doc] Adding docs for Kuberay KAI scheduler integration

[EkinKarabulut]

Why are these changes needed?

Adding the docs for kuberay KAI-Scheduler integration PR.
ray-project/kuberay#3886

Related issue number

ray-project/kuberay#3886

Checks

• I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
• N/A I've run scripts/format.sh to lint the changes in this PR.
• I've included any doc changes needed for https://docs.ray.io/en/master/.
  • N/A I've added any new APIs to the API Reference. For example, if I added a
    method in Tune, I've added it in doc/source/tune/api/ under the
    corresponding .rst file.
• I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
• Testing Strategy
  • Doc tests

[gemini-code-assist]

Summary of Changes

Hello @EkinKarabulut, 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 new documentation to guide users through integrating KubeRay with the KAI Scheduler. The aim is to empower users with advanced Kubernetes scheduling capabilities for their Ray clusters, focusing on efficient resource management, workload prioritization, and optimized GPU utilization.

Highlights

• New Documentation Added: I've added comprehensive documentation detailing the integration of KubeRay with NVIDIA's KAI Scheduler. This new guide provides users with instructions on how to leverage KAI Scheduler's advanced features for Ray clusters on Kubernetes.
• KAI Scheduler Capabilities: The documentation covers key KAI Scheduler functionalities, including gang scheduling (ensuring all Ray cluster components are scheduled together), hierarchical queue management with quotas and priorities for resource allocation, and fractional GPU sharing to maximize GPU utilization.
• Practical Implementation Guide: The new guide includes step-by-step instructions for installing KAI Scheduler and configuring the KubeRay operator to use it. It also provides practical YAML examples for creating KAI Scheduler queues, applying gang scheduling to RayClusters, setting workload priorities, and demonstrating GPU sharing for Ray workers.

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 or fill out our survey to provide feedback.

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 adds new documentation for integrating KubeRay with the KAI Scheduler. The new page provides a good overview and examples. My review focuses on improving the correctness and clarity of the code snippets and instructions to ensure users can follow them without issues. I've identified some critical and high-severity issues where commands would fail or configurations are incorrect, along with several medium-severity suggestions to improve the overall quality of the documentation.

[gemini-code-assist]

The user is instructed to run kubectl apply -f ray-cluster.kai-scheduler.yaml below, but there's no instruction to save the preceding YAML definition into that file. This can be confusing for users. Please add a note to save the YAML content to a file.

[gemini-code-assist]

Similar to a previous comment, the user is instructed to run kubectl apply below, but there's no instruction to save the YAML definition into a file. Please add a note to save the YAML content to a file (e.g., ray-cluster.kai-gpu-sharing.yaml).

[jjyao]

Will review and merge this after the kuberay PR is merged.

[github-actions]

This pull request has been automatically marked as stale because it has not had
any activity for 14 days. It will be closed in another 14 days if no further activity occurs.
Thank you for your contributions.

You can always ask for help on our discussion forum or Ray's public slack channel.

If you'd like to keep this open, just leave any comment, and the stale label will be removed.

[angelinalg]

Some style nits that we would appreciate you addressing. Generally we like to avoid using passive voice for clarity. Thank you for adding to the documentation and apologies for the delay.

[angelinalg]

Suggested change

	- **GPU Sharing**: Allow multiple Ray workloads from across teams to be packed on the same GPU, letting your organization fit more work onto your existing hardware and reducing idle GPU time.
	- **GPU sharing**: Allow Ray to pack multiple workloads from across teams on the same GPU, letting your organization fit more work onto your existing hardware and reducing idle GPU time.

[Future-Outlier]

Hi, @EkinKarabulut
do you have time contribute this?

[Future-Outlier]

cc @fscnick for review together, thank you!

[fscnick]

Thanks for addressing the feedback. LGTM

cc @Future-Outlier