CapacityLimit - Amazon Batch
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).

CapacityLimit

Defines the type and maximum quantity of resources that can be allocated to service jobs in a service environment.

Contents

capacityUnit

The unit of measure for the capacity limit, which defines how maxCapacity is interpreted. For SAGEMAKER_TRAINING jobs in a quota management enabled service environment, specify the instance type (for example, ml.m5.large). Otherwise, use NUM_INSTANCES.

Type: String

Required: No

maxCapacity

The maximum capacity available for the service environment. For a quota management enabled service environment, this value represents the maximum quantity of a particular resource type (specified by capacityUnit) that can be allocated to service jobs. For other service environments, this value represents the maximum quantity of all resources that can be allocated to service jobs.

For example, if maxCapacity=50 and capacityUnit=NUM_INSTANCES, you can run up to 50 instances concurrently. If you run 5 SageMaker Training jobs that each use 10 instances, a subsequent job requiring 10 instances waits in the queue until capacity is available. In a quota management enabled service environment with capacityUnit=ml.m5.large, only ml.m5.large instances count against this limit, and jobs requiring other instance types wait until a matching capacity limit is configured.

Type: Integer

Required: No

See Also

For more information about using this API in one of the language-specific Amazon SDKs, see the following: