example-full.yaml

# An unique identifier for the head node and workers of this cluster.
cluster_name: xwjiang-test
# The maximum number of workers nodes to launch in addition to the head
# node. This takes precedence over min_workers. min_workers default to 0.
min_workers: 3
initial_workers: 3
max_workers: 3
# autoscaling_mode: aggressive
# target_utilization_fraction: 0.9
# If a node is idle for this many minutes, it will be removed.
idle_timeout_minutes: 15
docker:
    image: "rayproject/ray:1.4.0-cpu" # You can change this to latest-cpu if you don't need GPU support and want a faster startup
    container_name: "ray_container"
    # If true, pulls latest version of image. Otherwise, `docker run` will only pull the image
    # if no cached version is present.
    pull_before_run: True
    run_options: []  # Extra options to pass into "docker run"
# Cloud-provider specific configuration.
provider:
    type: aws
    region: us-west-2
    availability_zone: us-west-2a
    cache_stopped_nodes: False
# How Ray will authenticate with newly launched nodes.
auth:
    ssh_user: ubuntu
    ssh_private_key: ~/Documents/aws_secrets/xwjiang-test.pem
head_node:
    InstanceType: m5ad.8xlarge
    ImageId: latest_dlami
    KeyName: xwjiang-test
    # InstanceMarketOptions:
    #     MarketType: spot
           # SpotOptions:
           #     MaxPrice: "9.0"
worker_nodes:
    InstanceType: g4dn.12xlarge 
    ImageId: latest_dlami
    KeyName: xwjiang-test
    # # Run workers on spot by default. Comment this out to use on-demand.
    # InstanceMarketOptions:
    #     MarketType: spot
        # SpotOptions:
        #     MaxPrice: "9.0"
file_mounts: {
    /root/tune/: /Users/xwjiang/ray/python/ray/tune/,
    /root/autoscaler/: /Users/xwjiang/ray/python/ray/autoscaler/,
    # "~/koch": "."
}
initialization_commands: []
setup_commands:
  - pip install -U -q boto3
  # - pip install -U ray==1.4.0 # https://s3-us-west-2.amazonaws.com/ray-wheels/latest/ray-2.0.0.dev0-cp37-cp37m-manylinux2014_x86_64.whl
  # - pip install -q ipdb ray[rllib] #torch torchvision
  # - rm -rf  /root/anaconda3/lib/python3.7/site-packages/ray/tune
  # - rm -rf  /root/anaconda3/lib/python3.7/site-packages/ray/autoscaler
  # - cp -r /root/tune/  /root/anaconda3/lib/python3.7/site-packages/ray/tune
  # - cp -r /root/autoscaler/  /root/anaconda3/lib/python3.7/site-packages/ray/autoscaler
  # - apt update && apt-get install -y libglib2.0-0 libgl1-mesa-glx
  # - pip install -U tensorflow
  - sudo apt update && sudo apt install -y build-essential
  - pip install statsmodels tqdm convertdate "pystan<3" fbprophet sklearn
# Custom commands that will be run on the head node after common setup.
head_setup_commands: []
# Custom commands that will be run on worker nodes after common setup.
worker_setup_commands: []
# # Command to start ray on the head node. You don't need to change this.
head_start_ray_commands:
    - ray stop
    - ray stop --force
    -  ray start --head --port=6379 --object-manager-port=8076 --autoscaling-config=~/ray_bootstrap_config.yaml --object-store-memory=10000000000
# Command to start ray on worker nodes. You don't need to change this.
worker_start_ray_commands:
    - ray stop
    - ray stop --force
    -  ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076 --object-store-memory=10000000000