Skip to content

Autoscaling configuration for custom endpoints misses some capabilities #305

@giuseppeporcelli

Description

@giuseppeporcelli

The autoscaling configuration for custom endpoints misses some configurations in the template, including min and max replica counts, cooldown period and more.
The Prometheus trigger is also missing.

Reference: https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-hyperpod-model-deployment-autoscaling.html#sagemaker-hyperpod-model-deployment-autoscaling-yaml

The problem is that these parameters are completely missing from the template so there is no way to use them with the current CLI.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions