Batch

Introduction

Batch is a cloud-based service provided by Amazon Web Services (AWS) that simplifies the process of running batch computing workloads on the AWS cloud infrastructure. Batch allows you to efficiently process large volumes of data and run batch jobs without the need to manage and provision underlying compute resources.

Under the hood, the local Docker engine is used to run the containers that simulate your Batch jobs.

LocalStack allows you to use the Batch APIs to automate and scale computational tasks in your local environment while handling batch workloads. Batch jobs are executed using the ECS runtime, allowing for support of managed compute environments and improved service compatibility.

The supported APIs are available on our API Coverage section, which provides information on the extent of Batch integration with LocalStack.

Getting started

This guide is designed for users new to AWS Batch and assumes basic knowledge of the AWS CLI and our awslocal wrapper script.

Start your LocalStack container using your preferred method. We will demonstrate how you create and run a Batch job by following these steps:

1. Creating a service role for the compute environment.
2. Creating the compute environment.
3. Creating a job queue using the compute environment.
4. Creating a job definition.
5. Submitting a job to the job queue.

Create a service role

You can create a role using the CreateRole API.

LocalStack requires the role to exist with a valid trust policy. When enforcing IAM policies, ensure that the policy is valid and the role is properly attached.

Run the following command to create a role for ECS task execution:

awslocal iam create-role \
    --role-name myrole \
    --assume-role-policy-document '{
  "Version": "2025-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Principal": {
        "Service": "ecs-tasks.amazonaws.com"
      },
      "Action": "sts:AssumeRole"
    }
  ]
}'

Then attach the ECS task execution policy:

awslocal iam attach-role-policy \
    --role-name myrole \
    --policy-arn arn:aws:iam::aws:policy/service-role/AmazonECSTaskExecutionRolePolicy

Create the compute environment

You can use the CreateComputeEnvironment API to create a compute environment.

Run the following command using the role ARN above (arn:aws:iam::000000000000:role/myrole) to create a managed compute environment with FARGATE:

awslocal batch create-compute-environment \
    --compute-environment-name myenv \
    --type MANAGED \
    --state ENABLED \
    --compute-resources type=FARGATE,maxvCpus=128,subnets=subnet-12345678,securityGroupIds=sg-12345678 \
    --service-role arn:aws:iam::000000000000:role/myrole

Note

While networking resources such as subnets and security groups are required as input, LocalStack does not create real cloud infrastructure. These values must still be present for the compute environment to be created.

Create a job queue

You can fetch the ARN using the DescribeComputeEnvironments API.

Run the following command to fetch the ARN of the compute environment:

awslocal batch describe-compute-environments --compute-environments myenv

Output

{
  "computeEnvironments": [
    {
      "computeEnvironmentName": "myenv",
      "computeEnvironmentArn": "arn:aws:batch:us-east-1:000000000000:compute-environment/myenv",
      "ecsClusterArn": "arn:aws:ecs:us-east-1:000000000000:cluster/OnDemand_Batch_abc123",
      "type": "MANAGED",
      "status": "VALID",
      "statusReason": "Compute environment is available",
      "serviceRole": "arn:aws:iam::000000000000:role/myrole"
    }
  ]
}

You can use the ARN to create the job queue using CreateJobQueue API.

Run the following command to create the job queue:

awslocal batch create-job-queue \
    --job-queue-name myqueue \
    --priority 1 \
    --compute-environment-order order=0,computeEnvironment=arn:aws:batch:us-east-1:000000000000:compute-environment/myenv \
    --state ENABLED

Create a job definition

Now, you can define what occurs during a job run. In this example, you can execute the ‘busybox’ container from DockerHub and initiate the command: ‘sleep 30’. It’s important to note you can override this command when submitting the job.

Run the following command to create the job definition using the RegisterJobDefinition API:

awslocal batch register-job-definition \
    --job-definition-name myjobdefn \
    --type container \
    --platform-capabilities FARGATE \
    --container-properties '{
        "image": "busybox",
        "resourceRequirements": [
            {"type": "VCPU", "value": "0.25"},
            {"type": "MEMORY", "value": "512"}
        ],
        "command": ["sleep", "30"],
        "networkConfiguration": {
            "assignPublicIp": "ENABLED"
        },
        "executionRoleArn": "arn:aws:iam::000000000000:role/myrole"
    }'

If you want to pass arguments to the command as parameters, you can use the Ref:: declaration to set placeholders for parameter substitution.

This allows the dynamic passing of values at runtime for specific job definitions.

awslocal batch register-job-definition \
    --job-definition-name myjobdefn \
    --type container \
    --parameters '{"time":"10"}' \
    --platform-capabilities FARGATE \
    --container-properties '{
        "image": "busybox",
        "resourceRequirements": [
            {"type": "VCPU", "value": "0.25"},
            {"type": "MEMORY", "value": "512"}
        ],
        "command": ["sleep", "Ref::time"],
        "networkConfiguration": {
            "assignPublicIp": "ENABLED"
        },
        "executionRoleArn": "arn:aws:iam::000000000000:role/myrole"
    }'

Submit a job to the job queue

You can now run a compute job. This command runs a job on the queue that you have set up previously, overriding the container command to run: sh -c "sleep 5; pwd". This command simulates work being done in the container.

Run the following command to submit a job to the job queue using the SubmitJob API:

awslocal batch submit-job \
    --job-name myjob \
    --job-queue myqueue \
    --job-definition myjobdefn \
    --container-overrides '{"command":["sh", "-c", "sleep 5; pwd"]}'

Current Limitations

LocalStack simulates the execution of ECS-based AWS Batch jobs using the local ECS runtime. No real infrastructure is created or managed.

Array jobs are supported in sequential mode only.

A subset of environment variables is supported, including:

• AWS_BATCH_CE_NAME
• AWS_BATCH_JOB_ARRAY_INDEX
• AWS_BATCH_JOB_ARRAY_SIZE
• AWS_BATCH_JOB_ATTEMPT
• AWS_BATCH_JOB_ID
• AWS_BATCH_JQ_NAME

The configuration variable ECS_DOCKER_FLAGS can be used to pass additional Docker flags to the container runtime.

Setting ECS_TASK_EXECUTOR=kubernetes is supported as an alternative backend, though Kubernetes execution is experimental and may not support all features.

API Coverage

Operation ▲	Implemented	Image

PreviousPage 1 of 0Next