Deploy Hadoop, Spark and Flink clusters on Amazon EC2 with this

cluster-deployment.yml

name: HadoopSparkFlinkCluster
ec2:
  region: us-west-2
  type: m3.xlarge
cookbooks:
  hadoop:
    branch: master
    github: karamel-lab/apache-hadoop-chef
  spark:
    branch: master
    github: ashansa/spark-chef
  flink:
    branch: master
    github: ashansa/flink-chef
attrs:
  hadoop:
    mr:
      user: ubuntu
    yarn:
      user: ubuntu
  hdfs:
    user: ubuntu
  spark:
    user: ubuntu
    driver_memory: 8192m
    executor_memory: 100g
  flink:
    user: ubuntu
    jobmanager:
      heap_mbs: '8192'
    taskmanager:
      heap_mbs: '102400'
groups:
  datanodes:
    size: 2
    ec2:
      region: us-west-2
      type: i2.4xlarge
      price: 0.5
    recipes:
    - spark::worker
    - hadoop::dn
    - hadoop::nm
    - flink::taskmanager
  namenodes:
    size: 1
    ec2:
      region: us-west-2
      type: m3.xlarge
      price: 0.15
    recipes:
    - flink::jobmanager
    - hadoop::nn
    - spark::master
    - hadoop::rm