Create a SageMaker Training job queue in AWS Batch

SageMaker Training job queues integrate directly with the SageMaker AI service to provide serverless job scheduling without requiring you to manage underlying compute infrastructure.

Prerequisites

Before creating a SageMaker Training job queue, ensure you have:

• Service environment – A service environment that defines capacity limits. For more information, see Create a service environment in AWS Batch.

• IAM permissions – Permissions to create and manage AWS Batch job queues and service environments. For more information, see AWS Batch IAM policies, roles, and permissions.

Create a SageMaker Training job queue (AWS Batch console)

1. Open the AWS Batch console at https://console.aws.amazon.com/batch/.

2. In the navigation pane, choose Job queues and the Create.

3. For Orchestration type, choose SageMaker Training.

4. For Job queue configuration:

   1. For Name, enter the name of the Job queue.

   2. for Priority, enter a value between 0 and 1000. A Job queue with a higher priority is given preference for service environments.

   3. (Optional) For Scheduling policy Amazon Resource Name (ARN), choose an existing scheduling policy.

   4. For Connected service environments, select a service environment from the list to associate with the job queue.

5. (Optional) For Job state limits:

   1. For Misconfiguration, choose SERVICE_ENVIRONMENT_MAX_RESOURCE and enter the Maximum runnable time (seconds).

   2. For Capacity, choose INSUFFICIENT_INSTANCE_CAPACITY and enter the Maximum runnable time (seconds).

6. Choose Create job queue

Create a SageMaker Training job queue (AWS CLI)

Use the create-job-queue command to create a SageMaker Training job queue.

The following example creates a basic SageMaker Training job queue that uses a service environment:

aws batch create-job-queue \
  --job-queue-name my-sm-training-fifo-jq \
  --job-queue-type SAGEMAKER_TRAINING \
  --priority 1 \
  --service-environment-order order=1,serviceEnvironment=ExampleServiceEnvironment

Replace ExampleServiceEnvironment with the name of your service environment.

The command returns output similar to the following:

{
  "jobQueueName": "my-sm-training-fifo-jq",
  "jobQueueArn": "arn:aws:batch:region:account:job-queue/my-sm-training-fifo-jq"
}

After creating your job queue, verify that it was created successfully and is in a valid state.

Use the describe-job-queues command to view details about your job queue:

aws batch describe-job-queues --job-queues my-sm-training-fifo-jq

The command returns output similar to the following:

{
  "jobQueues": [
    {
      "jobQueueName": "my-sm-training-fifo-jq",
      "jobQueueArn": "arn:aws:batch:region:account:job-queue/my-sm-training-fifo-jq",
      "state": "ENABLED",
      "status": "VALID",
      "statusReason": "JobQueue Healthy",
      "priority": 1,
      "computeEnvironmentOrder": [],
      "serviceEnvironmentOrder": [
        {
          "order": 1,
          "serviceEnvironment": "arn:aws:batch:region:account:service-environment/ExampleServiceEnvironment"
        }
      ],
      "jobQueueType": "SAGEMAKER_TRAINING",
      "tags": {},
      "jobStateTimeLimitActions": []
    }
  ]
}

Ensure that:

• The state is ENABLED

• The status is VALID

• The statusReason is JobQueue Healthy

• The jobQueueType is SAGEMAKER_TRAINING

• The serviceEnvironmentOrder references your service environment