Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add creation of kueue resources in e2e tests #616

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

ChughShilpa
Copy link

@ChughShilpa ChughShilpa commented Sep 26, 2024

Issue link

RHOAIENG-11047

What changes have been made

Since we are adjusting e2e tests to run with AMD gpu, thus

  • Adding creation of kueue resources as part of test to make it compatible with gpuresources (which is covered in next PR)
  • Removing creation of kueue resources from setup script
  • Adjust methods with updated Ray cluster client

Verification steps

Run test manually
go test -timeout 30m -v ./test/e2e -run TestMnistPyTorchAppWrapperCpu

Checks

  • I've made sure the tests are passing.
  • Testing Strategy
    • Unit tests
    • Manual tests
    • Testing is not required for this change

Copy link

openshift-ci bot commented Sep 26, 2024

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign christianzaccaria for approval. For more information see the Kubernetes Code Review Process.

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

Name: corev1.ResourceMemory,
NominalQuota: resource.MustParse("12Gi"),
},
{
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should it be removed if numberOfGpus is 0?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

IIRC specifying the resource name is mandatory even if nominalquota is 0

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The cleanest solution could be to move creation of ResourceFlavor and ClusterQueue (in the end also RayCluster) into Test<something>Gpu/Cpu functions. Each test function would then set properties relevant just to the test function.
That way we would get rid of conditions in test, reducing its reading complexity. Though that would mean a large rewrite.
I am ok with current solution too, just thinking out loud.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

IIRC specifying the resource name is mandatory even if nominalquota is 0

I think it is (was?) mandatory in RHOAI due to SDK passing GPU values. Not sure if it is still needed.
Anyway in real use cases for finetuning users need GPU.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Right, I'm thinking the same, it's an issue at the SDK level right? It should not request 0 GPU?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Created a jira issue for SDK to address this - https://issues.redhat.com/browse/RHOAIENG-13618
@astefanutti @sutaakar

test/e2e/mnist_rayjob_raycluster_test.go Outdated Show resolved Hide resolved
test/e2e/mnist_pytorch_appwrapper_test.go Show resolved Hide resolved
Copy link
Contributor

@sutaakar sutaakar left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

/lgtm

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants