Support network topology constrain with highestTierName

[zhaoqi612]

What type of PR is this?

/kind feature

What this PR does / why we need it:

When submitting a job and specifying the affinity tier, users can choose either of the two methods: highestTierName or highestTierAllowed, which makes the operation more convenient.

Which issue(s) this PR fixes:

Partial fixes for #4188

Special notes for your reviewer:

@JesseStutler @wangyang0616

Example YAML for vcJob after modification

apiVersion: batch.volcano.sh/v1alpha1
kind: Job
metadata:
  name: network-topology-job
spec:
  minAvailable: 6
  schedulerName: volcano
  networkTopology:
    mode: hard
    highestTierName: "volcano.sh/hypernode"
  tasks:
    - replicas: 6
      name: "task"
      partitionPolicy:
        totalPartitions: 2
        partitionSize: 3
        minPartitions: 2
        networkTopology:
            mode: hard
            highestTierName: "volcano.sh/hypernode"
      template:
        metadata:
          name: task
        spec:
          containers:
            - image: ubuntu
              imagePullPolicy: IfNotPresent
              name: task
              resources:
                requests:
                  cpu: "2"
                  memory: 2Gi
          restartPolicy: OnFailure

Example YAML for podgroup after modification

apiVersion: scheduling.volcano.sh/v1beta1
kind: PodGroup
metadata:
  name: network-topology-podgroup
spec:
  minMember: 6
  networkTopology:
    mode: hard 
    highestTierName: "volcano.sh/hypercluster"
  subGroupPolicy: 
    - subGroupSize: 3
      minSubGroups: 2
      name: task
      labelSelector:
        matchLabels:
           volcano.sh/task-spec: task
      matchLabelKeys:
        - volcano.sh/partition-id
      networkTopology:
        mode: hard 
        highestTierName: "volcano.sh/hypernode"

[volcano-sh-bot]

@zhaoqi612: The label(s) kind/what, kind/this, kind/pr, kind/does, kind//, kind/why, kind/we, kind/need, kind/it: cannot be applied, because the repository doesn't have them.

Details

In response to this:

What type of PR is this?

/kind

What this PR does / why we need it:

When submitting a job and specifying the affinity tier, users can choose either of the two methods: highestTierName or highestTierAllowed, which makes the operation more convenient.

Which issue(s) this PR fixes:

Fixes #4188

Special notes for your reviewer:

@JesseStutler @wangyang0616

Example YAML for vcJob after modification

apiVersion: batch.volcano.sh/v1alpha1
kind: Job
metadata:
 name: network-topology-job
spec:
 minAvailable: 6
 schedulerName: volcano
 networkTopology:
   mode: hard
   highestTierName: "volcano.sh/hypernode"
 tasks:
   - replicas: 6
     name: "task"
     partitionPolicy:
       totalPartitions: 2
       partitionSize: 3
       minPartitions: 2
       networkTopology:
           mode: hard
           highestTierName: "volcano.sh/hypernode"
     template:
       metadata:
         name: task
       spec:
         containers:
           - image: ubuntu
             imagePullPolicy: IfNotPresent
             name: task
             resources:
               requests:
                 cpu: "2"
                 memory: 2Gi
         restartPolicy: OnFailure

Example YAML for podgroup after modification

apiVersion: scheduling.volcano.sh/v1beta1
kind: PodGroup
metadata:
 name: network-topology-podgroup
spec:
 minMember: 6
 networkTopology:
   mode: hard 
   highestTierName: "volcano.sh/hypercluster"
 subGroupPolicy: 
   - subGroupSize: 3
     minSubGroups: 2
     name: task
     matchPolicy:
       - labelKey: volcano.sh/task-subgroup-id
     networkTopology:
       mode: hard 
       highestTierName: "volcano.sh/hypernode"

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.

[gemini-code-assist]

Summary of Changes

Hello @zhaoqi612, 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 improves the flexibility of network topology constraints within Volcano by introducing the highestTierName field. This allows users to define job affinity based on a more descriptive named tier, making configurations more intuitive. The changes span across API definitions, controller logic for job and podgroup management, and the core scheduler, ensuring that these new constraints are properly validated, processed, and applied during resource allocation. This enhancement provides a more user-friendly and robust mechanism for specifying network-aware scheduling requirements.

Highlights

• Enhanced Network Topology Constraints: Introduced the highestTierName field, allowing users to specify network topology affinity using a named tier (e.g., "volcano.sh/hypernode") instead of or in addition to the numerical highestTierAllowed.
• Mutual Exclusivity Validation: Added validation to ensure that users cannot specify both highestTierAllowed and highestTierName simultaneously within NetworkTopologySpec for both Jobs and PodGroups, preventing ambiguous configurations.
• Scheduler Integration: The scheduler now processes highestTierName by mapping it to its corresponding numerical tier during the allocation phase, ensuring that jobs with named tier constraints are correctly scheduled within the defined network topology.
• API and Internal Data Structure Updates: Modified internal data structures like HyperNodeInfo and ClusterInfo to store and manage tierName, along with new utility functions to build and update HyperNodes with this new field.
• Comprehensive Testing: New test cases have been added across controllers, schedulers, and webhooks to validate the correct behavior and error handling of the highestTierName functionality under various scenarios.

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 introduces the concept of tierName for network topology, allowing users to specify a symbolic name for a tier instead of a numerical highestTierAllowed. The changes involve adding a tierName field to HyperNodeInfo and related structures, updating utility functions to handle this new field, and modifying the job controller and scheduler to process highestTierName. Specifically, the scheduler now translates highestTierName to its corresponding numerical tier during allocation. Validation logic has been added to prevent simultaneous specification of highestTierAllowed and highestTierName in job and podgroup network topology configurations. Review comments highlighted a potential non-deterministic behavior in HyperNodeTierNameMap() if conflicting tierName to tier mappings exist, suggesting a warning log and consistent tier selection (e.g., minimum tier). Another comment pointed out that the job controller's createOrUpdatePodGroup function introduced a default HighestTierAllowed of 1 when networkTopology was specified without either HighestTierAllowed or HighestTierName, which could be a breaking change and was advised against for backward compatibility. Additionally, the error messages in validateNetworkTopology were improved for clarity and specificity.

[wangyang0616]

/lgtm

[wangyang0616]

defer to @JesseStutler

[wangyang0616]

For the matchPolicy field in SubgroupPolicy, the master branch has been updated. The PR description needs to be refreshed to reflect the latest definition. For details on the API changes, please refer to #202 #4785 .

[zhaoqi612]

For the matchPolicy field in SubgroupPolicy, the master branch has been updated. The PR description needs to be refreshed to reflect the latest definition. For details on the API changes, please refer to #202 #4785 .

[zhaoqi612]

For the matchPolicy field in SubgroupPolicy, the master branch has been updated. The PR description needs to be refreshed to reflect the latest definition. For details on the API changes, please refer to #202 #4785 .

done

[wangyang0616]

/lgtm
/approve

[volcano-sh-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: wangyang0616

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• pkg/controllers/OWNERS [wangyang0616]
• pkg/scheduler/OWNERS [wangyang0616]
• pkg/webhooks/OWNERS [wangyang0616]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment