View a markdown version of this page

AWS::SageMaker::InferenceComponent InferenceComponentAvailabilityZoneBalance - Amazon CloudFormation
SyntaxProperties
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

This is the new Amazon CloudFormation Template Reference Guide. Please update your bookmarks and links. For help getting started with CloudFormation, see the Amazon CloudFormation User Guide.

AWS::SageMaker::InferenceComponent InferenceComponentAvailabilityZoneBalance

Configuration for balancing inference component copies across Availability Zones.

Syntax

To declare this entity in your Amazon CloudFormation template, use the following syntax:

JSON

{
  "EnforcementMode" : String,
  "MaxImbalance" : Integer
}

YAML

  EnforcementMode: String
  MaxImbalance: Integer

Properties

EnforcementMode

Determines how strictly the Availability Zone balance constraint is enforced.

PERMISSIVE

The endpoint attempts to balance copies across Availability Zones but proceeds with scheduling even if balance can't be achieved due to available capacity or instance distribution across Availability Zones.

Required: Yes

Type: String

Allowed values: PERMISSIVE

Update requires: No interruption

MaxImbalance

The maximum allowed difference in the number of inference component copies between any two Availability Zones. This parameter applies only when the endpoint has instances across two or more Availability Zones. A copy placement is allowed if it reduces imbalance or the resulting imbalance is within this value.

Default value: 0.

Required: No

Type: Integer

Minimum: 0

Maximum: 100

Update requires: No interruption