Skip to content

Conversation

mburke5678
Copy link
Contributor

@mburke5678 mburke5678 commented Sep 25, 2025

https://issues.redhat.com/browse/OSDOCS-16432

Link to docs preview:

QE review:

  • QE has approved this change.

@mburke5678 mburke5678 added this to the Planned for 4.20 GA milestone Sep 25, 2025
@openshift-ci openshift-ci bot added the size/L Denotes a PR that changes 100-499 lines, ignoring generated files. label Sep 25, 2025
@ocpdocs-previewbot
Copy link

ocpdocs-previewbot commented Sep 25, 2025

🤖 Mon Oct 06 20:05:24 - Prow CI generated the docs preview:
https://99730--ocpdocs-pr.netlify.app
Complete list of updated preview URLs: artifacts/updated_preview_urls.txt

@mburke5678 mburke5678 changed the title Enable Dynamic Resource Allocations for openshift OSDOCS 12580 Enable Dynamic Resource Allocations for openshift Sep 29, 2025
@mburke5678
Copy link
Contributor Author

@tkashem Can you PTAL?

@openshift openshift deleted a comment from ocpdocs-vale-bot Oct 1, 2025
@openshift openshift deleted a comment from ocpdocs-vale-bot Oct 1, 2025
// Formerly out-of-cluster layering
:image-mode-os-out-caps: Out-of-cluster image mode
:image-mode-os-out-lower: out-of-cluster image mode
:attribute-based-full: Attribute-Based GPU Allocation in OpenShift with the NVIDIA GPU Operator
Copy link

Choose a reason for hiding this comment

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

we should probably omit with the NVIDIA GPU Operator, in 4.20 we are tech previewing Attribute-Based GPU Allocation only so vendors/partners can implement their DRA driver?


// Taken from https://issues.redhat.com/browse/OCPSTRAT-1756
{attribute-based-full} enables pods to request GPUs based on specific device attributes. This attribute-based resource allocation is achieved by the NVIDIA GPU Operator with a DRA driver, such as the NVIDIA Kubernetes DRA driver. This ensures that each pod receives the exact GPU specifications it requires.

Copy link

Choose a reason for hiding this comment

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

we don't support Nvidia GPU allocation in 4.20, so we should not probably mention Nvidia or their GPUs


Device class::
A device class is a category of devices that pods can claimed and how to select specific device attributes in claims. Some device drivers contain their own device class, such as the gpu.nvidia.com. Alternatively, an administrator can create device classes. A device class contains a device selector, which is a link:https://cel.dev/[common expression language (CEL)] expression that must evaluate to true if a device satisfies the request.
+
Copy link

Choose a reason for hiding this comment

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

the following specs refer to Nvidia GPU, oh so this will be marked as tech preview doc only?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

The tech preview notice is in the top file (the assembly file). (I made the name change, but haven't committed the changes yet.)

Copy link

openshift-ci bot commented Oct 6, 2025

@mburke5678: all tests passed!

Full PR test history. Your PR dashboard.

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-sigs/prow repository. I understand the commands that are listed here.

@mburke5678
Copy link
Contributor Author

@asahay19 PTAL

@asahay19
Copy link

asahay19 commented Oct 9, 2025

/lgtm

@openshift-ci openshift-ci bot added the lgtm Indicates that a PR is ready to be merged. label Oct 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
branch/enterprise-4.20 lgtm Indicates that a PR is ready to be merged. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants