Skip to content

add basic scale to zero notebook#47

Open
dulayjm wants to merge 1 commit intoaws-samples:mainfrom
dulayjm:add-normal-endpoint-scale-to-zero-docs
Open

add basic scale to zero notebook#47
dulayjm wants to merge 1 commit intoaws-samples:mainfrom
dulayjm:add-normal-endpoint-scale-to-zero-docs

Conversation

@dulayjm
Copy link

@dulayjm dulayjm commented Feb 22, 2025

Issue #, if available:

Description of changes:

Core Issue

The “scale-to-zero” autoscaling ManagedInstanceScaling introduced in the AWS blog does not work on endpoint models. Rather, it integrates only with inference components which are a new and separate abstraction from the rest of the SageMaker SDK. This leads to:

  • No integration with traditional SageMaker inference endpoints
  • No automatic integration with CloudWatch
  • No automatic integration with auto scaling
  • Broken integration with the AWS SageMaker Console.

Solution Overview

  • Create an endpoint config and endpoint without a model
  • Create a model without deploying to endpoint
  • Create instance component from the model
  • Attach the instance component to the endpoint
  • Manually configure CloudWatch and autoscaling

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant