Introducing the AWS Controllers for Kubernetes (ACK) | Amazon Web Services

We took an multi-phased approach, yielding hybrid custom+controller-runtimes:The above artifacts—Go code, container images, Kubernetes manifests for CRDs, roles, deployments, etc.—represent the business logic of how to manage AWS resources from within Kubernetes and are the responsibility of AWS service teams to create and maintain along with input from the community.To use ACK in a cluster you install the desired AWS service controller(s), considering that:As per the AWS Shared Responsibility Model, in the context of the cluster administration, you are responsible for regularly upgrading the ACK service controllers as well as applying security patches as they are made available.As an application developer, you create a namespaced custom resource in one of your ACK-enabled clusters. For example, let’s say you want ACK to create an Amazon Elastic Container Registry (ECR) repository, you’d define and subsequently apply something like:ACK service controllers installed by cluster admins can create, update, or delete AWS resources, based on the intent found in the custom resource defined in the previous step, by developers.