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

Test custom AMI for EFA support #8623

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

Conversation

mikkeloscar
Copy link
Contributor

@mikkeloscar mikkeloscar commented Dec 10, 2024

This PR is purely for enabling testing of an in-progress AMI to enable support for EFA on g5 instances.

The idea is that a cluster switched to this PR branch will be able to run a node-pool like this:

- config_items:
    labels: dedicated=gpu-g5-efa,zalando.org/nvidia-gpu=nvidia-a10g
    taints: dedicated=gpu-g5-efa:NoSchedule
    kuberuntu_distro_worker: "jammy_with_gpu"
    tag_instance_infrastructure_component: "false"
  discount_strategy: none
  instance_types:
  - g5.8xlarge
  - g5.12xlarge
  - g5.16xlarge
  - g5.24xlarge
  - g5.48xlarge
  max_size: 30
  min_size: 0
  name: gpu-g5-efa
  profile: worker-karpenter

With will run the special AMI defined as: kuberuntu_image_v1_31_jammy_with_gpu_amd64. To change the AMI to test, only that config-item needs to be changed in this PR and the node pool will be updated.

@mikkeloscar mikkeloscar added do-not-merge major Major feature changes or updates, e.g. feature rollout to a new country, new API calls. labels Dec 10, 2024
@mikkeloscar mikkeloscar force-pushed the datalab-test-efa-testing branch 3 times, most recently from eb8e7b2 to f364d61 Compare December 11, 2024 11:46
Signed-off-by: Mikkel Oscar Lyderik Larsen <[email protected]>
@mikkeloscar mikkeloscar force-pushed the datalab-test-efa-testing branch from f364d61 to c25d2fa Compare December 11, 2024 11:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
do-not-merge major Major feature changes or updates, e.g. feature rollout to a new country, new API calls.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant