CDK Orchestration for SageMaker Endpoint Auto Scaling Updates
Understand the story
The AWS Machine Learning Blog published a post on October 24, 2026, explaining how to use AWS CDK to orchestrate auto scaling for SageMaker AI real-time endpoints. The article highlights a practical deployment issue: when updating a SageMaker AI real-time endpoint's configuration, a registered scalable target can cause CloudFormation or AWS CDK deployments to fail; meanwhile, manually deregistering the scalable target may lead to rollback failures (UPDATE_ROLLBACK_FAILED). The article provides a solution for this update and rollback scenario. Only one report exists so far, and the event centers on the conflict between CDK/CloudFormation and SageMaker endpoint auto scaling resources during updates.
AI overview of reporting · Updated 12 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- AWS Machine Learning BlogOrchestrating SageMaker AI endpoint auto scaling with AWS CDK
When updating a SageMaker AI real-time endpoint configuration, the registered scalable target can cause CloudFormation or AWS CDK deployments to fail, and manually deregistering it may then lead to rollback failures (UPDATE_ROLLBACK_FAILED).
Story heat trend
Current heat7·Comparable peak10（10/24 01:00）·Comparable change over 24 hours –
The trend compares the same continuously observed sources, which may cover fewer sources than the current heat score. Hover or select the chart to inspect each hour; use the left and right arrow keys to navigate.