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

Proposal: Adoption of Feast to Kubeflow #807

Open
wants to merge 4 commits into
base: master
Choose a base branch
from

Conversation

franciscojavierarceo
Copy link
Contributor

@franciscojavierarceo franciscojavierarceo commented Jan 28, 2025

Feast has a long history with Kubeflow, as an add-on and previously included in the manifest dating back to kubeflow/manifests#1755.

After discussing with the https://github.com/feast-dev maintainers and getting their agreement, I am proposing donating Feast to Kubeflow to officially serve as Kubeflow's recommended open source feature store of choice.

Google document here for reference.

Related GitHub Issue Ref: #804

Signed-off-by: Francisco Javier Arceo <[email protected]>
Signed-off-by: Francisco Javier Arceo <[email protected]>
Signed-off-by: Francisco Javier Arceo <[email protected]>
Copy link

[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 andreyvelich 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

@franciscojavierarceo
Copy link
Contributor Author

@kubeflow/wg-data-leads @kubeflow/kubeflow-steering-committee @kubeflow/wg-pipeline-leads

@franciscojavierarceo
Copy link
Contributor Author

any thoughts / feedback @andreyvelich @terrytangyuan @johnugeorge @juliusvonkohout

@juliusvonkohout
Copy link
Member

I only added one comment and i am more worried about a secure multi-tenancy implementation and the storage layer than the general Idea. I think Kubeflow is a good place for feast.

Signed-off-by: Francisco Javier Arceo <[email protected]>
@indemnifyai
Copy link

I am +1 to the effort to add Feast to Kubeflow. Feast has proven to be a sustainable community. We do want the Kubeflow components to potentially have integrated workflows if valuable. A roadmap or strategic view on potential integrations is helpful. Should Feast be install by default in all Kubeflow deployments in the base manifest ? How will Feast be integration (or not) with the Dashboard. Finally, with all added projects, we should consider how we might turn down the project if contributions and users migrate away i.e. what would those metrics be and how that would happen. For example, if we don't have contributors and are not fixing bugs, do we keep installing Feast by default in the next release of the manifest. I am not saying this will happen but it will help future contributors on how to support Feast and other projects. One last point, the CNCF might want some statement on code history/quality i.e. license, CVEs, etc. We should some statement on that.

@Rostifar
Copy link

Rostifar commented Feb 4, 2025

We're using Feast at Affirm and seeing a native integration across the Kubeflow components would be great. I think adding Feast to Kubeflow is a great idea!

@shuchu
Copy link

shuchu commented Feb 11, 2025

We are using Kubeflow at Capital One and seeing a demand for a native Feature Store which can have seamless integration with Kuebflow, especially for the training part. I hope this integration can help the Kubeflow users to have a better experience to manage their features.

As a contributor of Feast, I will work closely with the Kubeflow contributors to make this integration move forward.

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.

5 participants