← Today's brief

Developers · Updated 23 Oct, 10:00 pm IST

AWS shows CDK pattern to auto-scale SageMaker endpoints without deployment failures

Image: AWS Machine Learning Blog

Why it matters for readers: Explains a technical reason why some SageMaker deployments fail and how AWS fixes it with CDK.

  • Updating SageMaker endpoint configuration fails when Application Auto Scaling targets remain registered because SageMaker rejects updates until targets are deregistered.1
  • CloudFormation cannot track Application Auto Scaling resources after you deregister scalable targets, which causes deployment failure modes.1
  • AWS provides a reusable CDK construct that uses CloudFormation custom resources to automatically deregister and re-register auto scaling during endpoint updates.1
  • With this approach you can change instance type, instance count, or environment variables and update autoscaling in a single deployment without manual steps.1

Get a brief like this every morning

Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.

Get started