Document NFD for GPU Labeling #44915

ArangoGutierrez · 2024-01-27T10:55:30Z

This patch documents the official k8s-sig NFD project for automated node labeling, and restructures the Automatic node labelling section at scheduling-gpus page

netlify · 2024-01-27T10:57:51Z

✅ Pull request preview available for checking

Built without sensitive environment variables

Name	Link
🔨 Latest commit	`07b14de`
🔍 Latest deploy log	https://app.netlify.com/sites/kubernetes-io-main-staging/deploys/65b92ba44707530008b71a00
😎 Deploy Preview	https://deploy-preview-44915--kubernetes-io-main-staging.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify site configuration.

ArangoGutierrez · 2024-01-27T10:57:58Z

/cc @marquiz

mikemckiernan

Nits and a (likely unwelcome) widening of the scope to identify some of the existing content as the manual method of labeling. I might be waltzing you into work that the doc maintainers don't want, so let's see what they say. Thanks as always for updating the docs!

content/en/docs/tasks/manage-gpus/scheduling-gpus.md

sftim · 2024-01-28T13:03:33Z

Thanks

/sig node

sftim

Hi; thanks for the pull request.

Here's my feedback on the changes.

content/en/docs/tasks/manage-gpus/scheduling-gpus.md

klueska · 2024-01-30T16:02:17Z

content/en/docs/tasks/manage-gpus/scheduling-gpus.md

+kind: Pod
+metadata:
+  name: example-vector-add
+spec:
+  # You can use Kubernetes node affinity to schedule this Pod onto a node
+  # that provides the kind of GPU that its container needs in order to work
+  affinity:
+    nodeAffinity:
+      requiredDuringSchedulingIgnoredDuringExecution:
+        nodeSelectorTerms:
+        - matchExpressions:
+          - key: "gpu.gpu-vendor.example/installed-memory"
+            operator: Gt # (greater than)
+            values: ["40535"]
+          - key: "gpu.gpu-vendor.example/family"
+            operator: In
+            values:
+            - Helium # example product family
+            - Neon   # example product family
+  restartPolicy: Never
+  containers:
+    - name: example-vector-add
+      image: "registry.example/example-vector-add:v42"
+      resources:
+        limits:
+          gpu-vendor.example/example-gpu: 1 # requesting 1 GPU
+{{< /highlight >}}


This seems to just be an extension of the example above, with the added affinity section, correct?

However, from reading the docs top-to-bottom, this implies that a label called gpu.gpu-vendor.example/installed-memorywill somehow magically be added to the node by NFD (which isn't true, an NFD plugin is needed to add this).

We should extend the text to make it clear that an NFD plugin is needed (in addition to NFD itself) in order to provide any vendor specific labels.

How about now?

content/en/docs/tasks/manage-gpus/scheduling-gpus.md

Signed-off-by: Carlos Eduardo Arango Gutierrez <eduardoa@nvidia.com>

sftim

This looks right.

Document NFD for GPU Labeling #44915 (review) added LGTM
Document NFD for GPU Labeling #44915 (comment) provided feedback
I judge that the feedback is addressed

It looks OK to publish this as a change to the live site.

/lgtm
/approve

Thanks everyone!

k8s-ci-robot · 2024-01-30T17:01:19Z

LGTM label has been added.

Git tree hash: f051228a1757be5c042884074869c5865761ae5d

k8s-ci-robot · 2024-01-30T17:01:20Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: eero-t, sftim

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

The pull request process is described here

Needs approval from an approver in each of these files:

~~content/en/OWNERS~~ [sftim]

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

content/en/docs/tasks/manage-gpus/scheduling-gpus.md

sftim · 2024-01-30T17:02:51Z

Actually I fixed some highlighting.

sftim · 2024-01-30T17:24:32Z

OK, it's less wrong now.
/lgtm

k8s-ci-robot · 2024-01-30T17:24:39Z

LGTM label has been added.

Git tree hash: f7cb87194f3090aa8dc8aac3251584f53886dc86

ArangoGutierrez · 2024-01-30T17:37:12Z

OK, it's less wrong now. /lgtm

Thanks a lot!

k8s-ci-robot added cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. language/en Issues or PRs related to English language labels Jan 27, 2024

k8s-ci-robot requested review from sftim and vishh January 27, 2024 10:55

k8s-ci-robot added the sig/docs Categorizes an issue or PR as relevant to SIG Docs. label Jan 27, 2024

ArangoGutierrez mentioned this pull request Jan 27, 2024

Document NFD in k/k kubernetes-sigs/node-feature-discovery#1569

Closed

k8s-ci-robot added the size/S Denotes a PR that changes 10-29 lines, ignoring generated files. label Jan 27, 2024

k8s-ci-robot requested a review from marquiz January 27, 2024 10:57

ArangoGutierrez force-pushed the automatedlabels branch from 1a1b976 to 8e84f8d Compare January 27, 2024 16:12

k8s-ci-robot added size/M Denotes a PR that changes 30-99 lines, ignoring generated files. and removed size/S Denotes a PR that changes 10-29 lines, ignoring generated files. labels Jan 27, 2024

ArangoGutierrez force-pushed the automatedlabels branch 3 times, most recently from e7e734c to b9cabb2 Compare January 27, 2024 16:39

mikemckiernan reviewed Jan 27, 2024

View reviewed changes

ArangoGutierrez force-pushed the automatedlabels branch from b9cabb2 to 4032470 Compare January 27, 2024 19:36

ArangoGutierrez requested a review from mikemckiernan January 27, 2024 19:39

k8s-ci-robot added the sig/node Categorizes an issue or PR as relevant to SIG Node. label Jan 28, 2024

bart0sh added this to Triage in SIG Node PR Triage Jan 28, 2024

bart0sh moved this from Triage to Needs Reviewer in SIG Node PR Triage Jan 28, 2024

sftim reviewed Jan 28, 2024

View reviewed changes

ArangoGutierrez force-pushed the automatedlabels branch from 4032470 to 31c34dc Compare January 29, 2024 10:11

ArangoGutierrez requested review from sftim and eero-t January 29, 2024 10:12

eero-t reviewed Jan 29, 2024

View reviewed changes

content/en/docs/tasks/manage-gpus/scheduling-gpus.md Outdated Show resolved Hide resolved

content/en/docs/tasks/manage-gpus/scheduling-gpus.md Outdated Show resolved Hide resolved

ArangoGutierrez force-pushed the automatedlabels branch from 31c34dc to 0d6fc88 Compare January 29, 2024 12:16

ArangoGutierrez requested a review from eero-t January 29, 2024 12:16

klueska reviewed Jan 30, 2024

View reviewed changes

sftim reviewed Jan 30, 2024

View reviewed changes

content/en/docs/tasks/manage-gpus/scheduling-gpus.md Outdated Show resolved Hide resolved

sftim reviewed Jan 30, 2024

View reviewed changes

content/en/docs/tasks/manage-gpus/scheduling-gpus.md Outdated Show resolved Hide resolved

ArangoGutierrez force-pushed the automatedlabels branch from 1e687dd to c8d9eed Compare January 30, 2024 16:17

k8s-ci-robot removed the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Jan 30, 2024

k8s-ci-robot requested review from reylejano and stmcginnis January 30, 2024 16:17

Document NFD for GPU Labeling

c4937ba

Signed-off-by: Carlos Eduardo Arango Gutierrez <eduardoa@nvidia.com>

ArangoGutierrez force-pushed the automatedlabels branch from c8d9eed to c4937ba Compare January 30, 2024 16:24

ArangoGutierrez requested review from klueska and sftim January 30, 2024 16:30

sftim reviewed Jan 30, 2024

View reviewed changes

k8s-ci-robot assigned sftim Jan 30, 2024

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Jan 30, 2024

k8s-ci-robot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Jan 30, 2024

sftim reviewed Jan 30, 2024

View reviewed changes

content/en/docs/tasks/manage-gpus/scheduling-gpus.md Outdated Show resolved Hide resolved

Fix highlighting

07b14de

k8s-ci-robot removed the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Jan 30, 2024

k8s-ci-robot requested a review from sftim January 30, 2024 17:02

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Jan 30, 2024

k8s-ci-robot merged commit 114fa30 into kubernetes:main Jan 30, 2024
6 checks passed

SIG Node PR Triage automation moved this from Needs Reviewer to Done Jan 30, 2024

sftim mentioned this pull request Jan 30, 2024

Fix highlighting for added lines #44954

Merged

windsonsea mentioned this pull request Jan 31, 2024

[zh] Add translations to scheduling-gpus.md #44961

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Document NFD for GPU Labeling #44915

Document NFD for GPU Labeling #44915

ArangoGutierrez commented Jan 27, 2024

netlify bot commented Jan 27, 2024 •

edited

Loading

ArangoGutierrez commented Jan 27, 2024

mikemckiernan left a comment

sftim commented Jan 28, 2024

sftim left a comment

klueska Jan 30, 2024

ArangoGutierrez Jan 30, 2024

sftim left a comment

k8s-ci-robot commented Jan 30, 2024

k8s-ci-robot commented Jan 30, 2024

sftim commented Jan 30, 2024

sftim commented Jan 30, 2024

k8s-ci-robot commented Jan 30, 2024

ArangoGutierrez commented Jan 30, 2024

Document NFD for GPU Labeling #44915

Document NFD for GPU Labeling #44915

Conversation

ArangoGutierrez commented Jan 27, 2024

netlify bot commented Jan 27, 2024 • edited Loading

✅ Pull request preview available for checking

ArangoGutierrez commented Jan 27, 2024

mikemckiernan left a comment

Choose a reason for hiding this comment

sftim commented Jan 28, 2024

sftim left a comment

Choose a reason for hiding this comment

klueska Jan 30, 2024

Choose a reason for hiding this comment

ArangoGutierrez Jan 30, 2024

Choose a reason for hiding this comment

sftim left a comment

Choose a reason for hiding this comment

k8s-ci-robot commented Jan 30, 2024

k8s-ci-robot commented Jan 30, 2024

sftim commented Jan 30, 2024

sftim commented Jan 30, 2024

k8s-ci-robot commented Jan 30, 2024

ArangoGutierrez commented Jan 30, 2024

netlify bot commented Jan 27, 2024 •

edited

Loading