containers-roadmap VS aws-eks-share-gpu

Thanks for showing up and answering questions. Congratulations on the release.

What kind of plans for support of Rust's evolving async ecosystem?

Any particular reason why the public roadmap does not show the columns similar to "Researching", "We're Working On It" like the other similar public AWS Roadmaps? See example for Containers: https://github.com/aws/containers-roadmap/projects/1

Would be nice to have fully working examples on Github, for most common scenarios across most AWS services. This is something that historically

Second, We will only rely on one managed node group, but we will leverage Karpenter; however, karpenter needs to be deployed on a node. (This may change soon once the Karpenter is available on the EKS Control Plane.) [EKS] Karpenter inside control plane · Issue #1792 · aws/containers-roadmap

As written in the AWS documentation, during periods of extremely high demand, Fargate Spot capacity might be unavailable. In concrete terms, if your ECS service is set up to execute tasks in 100% Spot, there is a risk of running out of capacity. A workaround has been created in the hope that one day this issue will be implemented by the AWS team. This workaround allows you to set up two ECS services :

Deploying Fargate with CDK has to have been the most pleasant developer experience I have ever had with any product so far.

If image caching becomes a reality with Fargate I can't imagine a need to ever use anything else

https://github.com/aws/containers-roadmap/issues/696

Eks Fargate doesn't support spot yet https://github.com/aws/containers-roadmap/issues/622

Until podman could be used with AWS ECR/ECS it's pretty much moot in my case: https://github.com/aws/containers-roadmap/issues/626

This project (available here) uses the k8s device plugin described by this AWS blog post to make GPU-based nodes publish the amount of GPU resource they have available. Instead of the amount of VRAM available or some abstract metric, this plugin advertises the amount of pods/processes that can be connected to the GPU. This is controlled by what is called by NVIDIA as Multi-Process Service (MPS).