Reference: AWS Batch job scenarios using EcsProperties
To illustrate how AWS Batch job definitions that use EcsProperties
can be structured based on your needs, this topic presents the following RegisterJobDefinition
payloads. You can copy these examples into a file, customize them to your needs, and then use the AWS Command Line Interface (AWS CLI) to call RegisterJobDefinition
.
AWS Batch job for Amazon ECS on Amazon EC2
The following is an example of a AWS Batch job for Amazon Elastic Container Service on Amazon Elastic Compute Cloud:
{ "jobDefinitionName": "multicontainer-ecs-ec2", "type": "container", "ecsProperties": { "taskProperties": [ { "containers": [ { "name": "c1", "essential": false, "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "2" }, { "type": "MEMORY", "value": "4096" } ] }, { "name": "c2", "essential": true, "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "6" }, { "type": "MEMORY", "value": "12288" } ] } ] } ] } }
AWS Batch job for Amazon ECS on Fargate
The following is an example of a AWS Batch job for Amazon Elastic Container Service on AWS Fargate:
{ "jobDefinitionName": "multicontainer-ecs-fargate", "type": "container", "platformCapabilities": [ "FARGATE" ], "ecsProperties": { "taskProperties": [ { "containers": [ { "name": "c1", "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "2" }, { "type": "MEMORY", "value": "4096" } ] }, { "name": "c2", "essential": true, "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "6" }, { "type": "MEMORY", "value": "12288" } ] } ], "executionRoleArn": "arn:aws:iam::1112223333:role/ecsTaskExecutionRole" } ] } }
AWS Batch job for Amazon EKS
The following is an example of a AWS Batch job for Amazon Elastic Kubernetes Service:
{ "jobDefinitionName": "multicontainer-eks", "type": "container", "eksProperties": { "podProperties": { "shareProcessNamespace": true, "initContainers": [ { "name": "init-container", "image": "public.ecr.aws/amazonlinux/amazonlinux:2", "command": [ "echo" ], "args": [ "hello world" ], "resources": { "requests": { "cpu": "1", "memory": "512Mi" } } }, { "name": "init-container-2", "image": "public.ecr.aws/amazonlinux/amazonlinux:2", "command": [ "echo", "my second init container" ], "resources": { "requests": { "cpu": "1", "memory": "512Mi" } } } ], "containers": [ { "name": "c1", "image": "public.ecr.aws/amazonlinux/amazonlinux:2", "command": [ "echo world" ], "resources": { "requests": { "cpu": "1", "memory": "512Mi" } } }, { "name": "sleep-container", "image": "public.ecr.aws/amazonlinux/amazonlinux:2", "command": [ "sleep", "20" ], "resources": { "requests": { "cpu": "1", "memory": "512Mi" } } } ] } } }
MNP AWS Batch job with multiple containers per node
The following is an example of a multi-node parallel (MNP) AWS Batch job with multiple containers per node:
{ "jobDefinitionName": "multicontainer-mnp", "type": "multinode", "nodeProperties": { "numNodes": 6, "mainNode": 0, "nodeRangeProperties": [ { "targetNodes": "0:5", "ecsProperties": { "taskProperties": [ { "containers": [ { "name": "range05-c1", "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "2" }, { "type": "MEMORY", "value": "4096" } ] }, { "name": "range05-c2", "command": [ "echo", "hello world" ], "image": "public.ecr.aws/amazonlinux/amazonlinux:latest", "resourceRequirements": [ { "type": "VCPU", "value": "2" }, { "type": "MEMORY", "value": "4096" } ] } ] } ] } } ] } }