search

kubernetes-sigs / cluster-api-provider-aws

Kubernetes Cluster API Provider AWS provides consistent deployment and day 2 operations of "self-managed" and EKS Kubernetes clusters on AWS.
http://cluster-api-aws.sigs.k8s.io/
Apache License 2.0
646 stars 575 forks source link

CapacityReservations/ Capacity Blocks for ML(GPU) in AWS to get allocation/ significant discounts. #5045

Open athiruma opened 4 months ago

athiruma commented 4 months ago

/kind feature

Describe the solution you'd like [A clear and concise description of what you want to happen.]

By using capacity blocks for ML, one can obtain a significant discount compared to on-demand GPU instances. However, we can also use CapacityReservations to allocate additional on-demand instances to the cluster in case of poor availability of on-demand instances.

Anything else you would like to add: [Miscellaneous information that will assist in solving the issue.] If the user uses the on-demand CapacityReservationId to the cluster if the reservation expires the cluster falls into normal on-demand instances. But for the Capacity Blocks, the instances will start deleted as these are GPU instances which need to be allocated to other users.

Environment:

k8s-ci-robot commented 4 months ago

This issue is currently awaiting triage.

If CAPA/CAPI contributors determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes-sigs/prow](https://github.com/kubernetes-sigs/prow/issues/new?title=Prow%20issue:) repository.