cluster-deployment.yml

name: StreamingSetup
ec2:
  region: us-west-2
  type: m3.medium
cookbooks:
  hadoop:
    branch: master
    github: karamel-lab/apache-hadoop-chef
  spark:
    branch: master
    github: karamel-lab/spark-chef
  flink:    
    branch: master
    github: karamel-lab/flink-chef
attrs:
  hadoop:
    mr:
      user: ubuntu
    yarn:
      user: ubuntu
    version: 2.7.1
  hdfs:
    user: ubuntu
  spark:
    user: ubuntu
    driver_memory: 8192m
    executor_memory: 30g
  flink:
    user: ubuntu
    mode: streaming
    jobmanager:
      heap_mbs: '8192'
    taskmanager:
      heap_mbs: '307200'
groups:
  namenodes:
    size: 1
    ec2:
      region: us-west-2
      type: m3.xlarge
      price: 0.05  
    recipes:
    - hadoop::nn
    - hadoop::rm
    - spark::master
    - flink::jobmanagerstreaming
  datanodes:
    size: 1
    ec2:
      region: us-west-2
      type: r2.2xlarge
      price: 0.05
    recipes:
    - hadoop::nm
    - hadoop::dn
    - flink::taskmanager
    - spark::worker