AWS CloudFormation template to create EMR-5.1.0 cluster

aws_emr.template.json

{
  "AWSTemplateFormatVersion": "2010-09-09",
  "Description": "Create AMR Cluster",
  "Parameters": {
    "ProductParam": {
      "Description": "The EMR Product tag",
      "Type": "String",
      "Default": "aProductParam",
      "MinLength": "3",
      "MaxLength": "50",
      "AllowedPattern": "^[a-zA-Z0-9][a-zA-Z0-9.-]*"
    },
    "ChargeCodeParam": {
      "Description": "The EMR ChargeCode tag",
      "Type": "String",
      "Default": "aChargeCodeParam",
      "MinLength": "3",
      "MaxLength": "50",
      "AllowedPattern": "^[a-zA-Z0-9][a-zA-Z0-9.-]*"
    },
    "EmrLogsBucketNameParam": {
      "Description": "The EMR logs S3 bucket name",
      "Type": "String",
      "Default": "emr-logs",
      "MinLength": "3",
      "MaxLength": "50",
      "AllowedPattern": "^[a-z0-9][a-z0-9.-]*"
    },
    "Ec2KeyNameParam":{
      "Description": "Name of an existing EC2 KeyPair to enable SSH access to the instance",
      "Type": "AWS::EC2::KeyPair::KeyName",
      "ConstraintDescription": "must be the name of an existing EC2 KeyPair."
    }
  },
  "Outputs": {
    "OutEMRTaskInstanceConfig": {
      "Value": {
        "Ref": "EMRTaskInstanceConfig"
      }
    },
    "OutEmrLogsS3Bucket": {
      "Value": {
        "Ref": "EmrLogsS3Bucket"
      }
    },
    "OutEMRClusterEC2RoleInstance": {
      "Value": {
        "Ref": "EMRClusterEC2RoleInstance"
      }
    }
  },
  "Resources": {
    "EmrLogsS3Bucket": {
      "Type": "AWS::S3::Bucket",
      "Properties": {
        "BucketName": {
          "Fn::Join": [
            "-",
            [
              {
                "Ref": "AWS::AccountId"
              },
              {
                "Ref": "EmrLogsBucketNameParam"
              }
            ]
          ]
        },
        "AccessControl": "Private"
      },
      "DeletionPolicy" : "Retain"
    },
    "EMRClusterEC2Role": {
      "Type": "AWS::IAM::Role",
      "Properties": {
        "AssumeRolePolicyDocument": {
          "Version": "2012-10-17",
          "Statement": [
            {
              "Effect": "Allow",
              "Action": "sts:AssumeRole",
              "Principal": {
                "Service": "ec2.amazonaws.com"
              }
            }
          ]
        },
        "ManagedPolicyArns": [
          "arn:aws:iam::aws:policy/service-role/AmazonElasticMapReduceforEC2Role"
        ],
        "RoleName": "EMRClusterEC2Role"
      }
    },
    "EMRClusterEC2RoleInstance": {
      "Type": "AWS::IAM::InstanceProfile",
      "Properties": {
        "Path": "/",
        "Roles": [
            {"Ref":"EMRClusterEC2Role"}
        ]
      },
      "DependsOn": [
        "EMRClusterEC2Role"
      ]
    },
    "EMRServiceRole": {
      "Type": "AWS::IAM::Role",
      "Properties": {
        "AssumeRolePolicyDocument": {
          "Version": "2012-10-17",
          "Statement": [
            {
              "Effect": "Allow",
              "Action": "sts:AssumeRole",
              "Principal": {
                "Service": "elasticmapreduce.amazonaws.com"
              }
            }
          ]
        },
        "ManagedPolicyArns": [
          "arn:aws:iam::aws:policy/service-role/AmazonElasticMapReduceRole"
        ],
        "RoleName": "EMRServiceRole"
      }
    },
    "EMRCluster": {
      "Type": "AWS::EMR::Cluster",
      "Properties": {
        "Name": "EMRCluster",
        "Instances": {
          "MasterInstanceGroup": {
            "InstanceCount": 1,
            "InstanceType": "m3.xlarge",
            "Market": "ON_DEMAND",
            "Name": "Master"
          },
          "CoreInstanceGroup": {
            "InstanceCount": 1,
            "InstanceType": "m3.xlarge",
            "Market": "ON_DEMAND",
            "Name": "Core"
          },
          "Ec2KeyName": {"Ref":"Ec2KeyNameParam"},
          "TerminationProtected": false
        },
        "ReleaseLabel": "emr-5.1.0",
        "Applications": [
          {
            "Name": "Spark"
          },
          {
            "Name": "Hadoop"
          },
          {
            "Name": "Hive"
          },
          {
            "Name": "Sqoop"
          }
        ],
        "Tags": [
          {
            "Key": "Product",
            "Value": {
              "Ref": "ProductParam"
            }
          },
          {
            "Key": "ChargeCode",
            "Value": {
              "Ref": "ChargeCodeParam"
            }
          }
        ],
        "LogUri": {
          "Fn::Join": [
            "",
            [
              "s3://",
              {
                "Ref": "EmrLogsS3Bucket"
              }
            ]
          ]
        },
        "JobFlowRole": {"Fn::GetAtt" : ["EMRClusterEC2RoleInstance", "Arn"] },
        "ServiceRole": "EMRServiceRole"
      },
      "DependsOn": [
        "EMRClusterEC2Role",
        "EMRServiceRole",
        "EMRClusterEC2RoleInstance",
        "EmrLogsS3Bucket"
      ]
    },
    "EMRTaskInstanceConfig": {
      "Type": "AWS::EMR::InstanceGroupConfig",
      "Properties": {
        "InstanceCount": 1,
        "InstanceType": "m3.xlarge",
        "InstanceRole": "TASK",
        "Market": "ON_DEMAND",
        "Name": "EMRTaskInstanceConfig",
        "JobFlowId": {
          "Ref": "EMRCluster"
        }
      },
      "DependsOn": [
        "EMRCluster"
      ]
    }
  }
}