View a markdown version of this page

Resource: Compute environment template - 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).

Resource: Compute environment template

The following example shows an empty compute environment template. You can use this template to create your compute environment that can then be saved to a file and used with the Amazon CLI --cli-input-json option. For more information about these parameters, see CreateComputeEnvironment in the Amazon Batch API Reference.

Before creating a managed Amazon EC2 compute environment, make sure you have the following prerequisites in place. These prerequisites apply when the type field is set to MANAGED.

  • Security group – Your compute resources require a security group that allows outbound traffic so that instances can communicate with the Amazon ECS service endpoint and pull container images. For more information, see Create a security group.

  • IAM roles – Amazon Batch requires an Amazon ECS instance role that allows container instances to make Amazon API calls on your behalf. For more information, see Amazon ECS instance role and Using service-linked roles for Amazon Batch.

    Note

    The instanceRole field accepts an instance profile ARN, not a role ARN. The format is arn:aws:iam::account_id:instance-profile/ecsInstanceRole.

  • Network access – Compute resources must be able to reach the Amazon ECS service endpoint. If your instances are in a private subnet without a public IP address, you can use either a NAT gateway or Amazon VPC interface endpoints. For more information, see Use an interface endpoint to Access Amazon Batch.

Note

You can generate a compute environment template with the following Amazon CLI command.

$ aws batch create-compute-environment --generate-cli-skeleton
Important

Compute environments must be created in ENABLED state.

The following example shows a skeleton template for a managed Amazon EC2 compute environment. The computeResources block is required when type is MANAGED.

{
    "computeEnvironmentName": "",
    "type": "MANAGED",
    "state": "ENABLED",
    "computeResources": {
        "type": "EC2",
        "allocationStrategy": "BEST_FIT_PROGRESSIVE",
        "minvCpus": 0,
        "maxvCpus": 16,
        "desiredvCpus": 0,
        "instanceTypes": [
            "default_arm64"
        ],
        "subnets": [
            "subnet-a1b2c3d4"
        ],
        "securityGroupIds": [
            "sg-a1b2c3d4"
        ],
        "instanceRole": "arn:aws:iam::123456789012:instance-profile/ecsInstanceRole",
        "tags": {
            "KeyName": ""
        },
        "launchTemplate": {
            "launchTemplateId": "",
            "version": "$Default"
        },
        "ec2Configuration": [
            {
                "imageType": "ECS_AL2023"
            }
        ]
    },
    "serviceRole": "",
    "tags": {
        "KeyName": ""
    }
}

The following example shows a skeleton template for an unmanaged Amazon EC2 compute environment. The computeResources block is not used for UNMANAGED compute environments and should be omitted.

{
    "computeEnvironmentName": "",
    "type": "UNMANAGED",
    "state": "ENABLED",
    "unmanagedvCpus": 0,
    "serviceRole": "",
    "tags": {
        "KeyName": ""
    }
}