Add support for HyperPod nodes

surajkota commented 1 month ago

Description of changes

SageMaker HyperPod recently launched EKS integration. This commit adds SageMaker instance types and toleration for running DeepHealthChecks so customers can install EFA helm chart without modifications unless required

Checklist

[x] Added/modified documentation as required (such as the README.md for modified charts)
[x] Incremented the chart version in Chart.yaml for the modified chart(s)
[x] Manually tested. Describe what testing was done in the testing section below
[x] Make sure the title of the PR is a good description that can go into the release notes

Testing

Installed EFA driver from my github branch and verified pods are scheduled on g5.8x instances for EC2 and Hyperpods without modifying the chart. Didnt run workloads as I didnt change the deamonset itself

helm install aws-efa-k8s-device-plugin .
NAME: aws-efa-k8s-device-plugin
LAST DEPLOYED: Wed Oct 23 09:23:44 2024
NAMESPACE: default
STATUS: deployed
REVISION: 1
TEST SUITE: None
NOTES:
EFA device plugin is installed, it can be requested as `vpc.amazonaws.com/efa` resource.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

bryantbiggs commented 1 month ago

partial dupe of #1129

surajkota commented 1 month ago

Ack, I can work with Nathan to close the other PR

surajkota commented 1 month ago

@bryantbiggs is there also a kustomize version of EFA device plugin maintained by EKS/AWS? or any other high traffic download sources where similar change should also go?

I checked the aws-samples repo and its depreciated in favor or this chart.

bryantbiggs commented 1 month ago

no - this is the source of truth for deploying the EFA device plugin into an EKS cluster

aws / eks-charts