GeoWave AWS

geowave.yaml

AWSTemplateFormatVersion: 2010-09-09

Parameters:
  Environment:
    Type: String
    Default: Production
    Description: The name of the enviornment this cluster is being deployed to.
  EmrVersion:
    Type: String
    Default: emr-5.29.0
    AllowedValues:
      - emr-5.29.0
      - emr-5.28.1
      - emr-5.28.0
      - emr-5.27.0
      - emr-5.26.0
    Description: The version of Amazon AWS EMR.
  GeoWaveVersion:
    Type: String
    Default: 1.1.0
    AllowedValues:
      - 1.1.0
      - 0.9.8
      - 0.9.7
      - 0.9.6
      - 0.9.5
      - 0.9.4
      - 0.9.3
    Description: >
      The version of GeoWave that will be deployed to the EMR cluster.
  S3Bucket:
    Type: String
    Description: >
      The bucket that will be used to persis ALL data associated with this cluster. This includes the
      HBase root folder, logs, and any associated Jupyter Notebooks.
  SubnetId:
    Type: AWS::EC2::Subnet::Id
    Description: The subnet that the EMR cluster will be deployed in.
  Ec2KeyName:
    Type: AWS::EC2::KeyPair::KeyName
    Description: >
      The name of the EC2 key pair that will be used to provision the master node.

Resources:

  # TODO: Performance tuning as per https://docs.aws.amazon.com/emr/latest/ReleaseGuide/emr-hbase-s3.html

  GeoWaveCluster:
    Type: AWS::EMR::Cluster
    Properties:
      Name: !Sub '${AWS::StackName}-${Environment}'
      JobFlowRole: !Ref EmrEc2InstanceProfile
      ServiceRole: !Ref EmrRole
      ReleaseLabel: !Ref EmrVersion
      LogUri: !Sub "s3://${S3Bucket}/logs"
      Applications:
        - Name: Hadoop
        - Name: Hue
        - Name: Hive
        - Name: Pig
        - Name: Spark
        - Name: HBase
        - Name: ZooKeeper
        - Name: Livy
        - Name: JupyterHub
      Configurations:
        - Classification: 'hbase-site'
          ConfigurationProperties:
            'hbase.rootdir': !Sub "s3://${S3Bucket}/hbase-root"
        - Classification: 'hbase'
          ConfigurationProperties:
            'hbase.emr.storageMode': 's3'
        - Classification: 'emrfs-site'
          ConfigurationProperties:
            'fs.s3.consistent': 'true'
        - Classification: 'jupyter-s3-conf'
          ConfigurationProperties:
            's3.persistence.enabled': 'true'
            's3.persistence.bucket': !Ref S3Bucket
        - Classification: 'hive-site'
          ConfigurationProperties:
            'hive.metastore.client.factory.class': 'com.amazonaws.glue.catalog.metastore.AWSGlueDataCatalogHiveClientFactory'
      BootstrapActions:
        - Name: GeoWave Bootstrap
          ScriptBootstrapAction:
              Path: !Sub 's3://geowave/${GeoWaveVersion}/scripts/emr/hbase/bootstrap-geowave.sh'
        - Name: GeoWave Jupyter Notebook Bootstrap
          ScriptBootstrapAction:
              Path: !Sub 's3://geowave/${GeoWaveVersion}/scripts/emr/jupyter/bootstrap-jupyter.sh'
      EbsRootVolumeSize: 32
      Instances:
        MasterInstanceFleet:
          Name: Master
          TargetOnDemandCapacity: 1
          InstanceTypeConfigs:
            - InstanceType: m5.xlarge
        CoreInstanceFleet:
          Name: Core
          TargetSpotCapacity: 1
          InstanceTypeConfigs:
            - InstanceType: m4.xlarge
              BidPriceAsPercentageOfOnDemandPrice: 35
        Ec2KeyName: !Ref Ec2KeyName
        Ec2SubnetId: !Ref SubnetId

  EmrRole:
    Type: AWS::IAM::Role
    Properties:
      AssumeRolePolicyDocument:
        Version: 2008-10-17
        Statement:
          - Effect: Allow
            Principal:
              Service: elasticmapreduce.amazonaws.com
            Action: 'sts:AssumeRole'
      Path: /
      ManagedPolicyArns:
        - 'arn:aws:iam::aws:policy/service-role/AmazonElasticMapReduceRole'
      Tags:
        - Key: Environment
          Value: !Ref Environment
  
  EmrEc2Role:
    Type: AWS::IAM::Role
    Properties:
      AssumeRolePolicyDocument:
        Version: '2012-10-17'
        Statement:
          - Effect: Allow
            Principal:
              Service: ec2.amazonaws.com
            Action: 'sts:AssumeRole'
      Path: /
      ManagedPolicyArns:
        - 'arn:aws:iam::aws:policy/service-role/AmazonElasticMapReduceforEC2Role'
        - 'arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore'
      Tags:
        - Key: Environment
          Value: !Ref Environment
  
  EmrEc2InstanceProfile:
    Type: AWS::IAM::InstanceProfile
    Properties:
      Path: /
      Roles:
        - !Ref EmrEc2Role

Outputs:
  GeoWaveCluster:
    Value: !Ref GeoWaveCluster
  EmrEc2Role:
    Value: !Ref EmrEc2Role