# Experimental: an example of configuring a mixed-node-type cluster.
cluster_name: alex
min_workers: 2
max_workers: 40

# Cloud-provider specific configuration.
provider:
    type: aws
    region: us-west-1
    availability_zone: us-west-1a

# Tell the autoscaler the allowed node types and the resources they provide.
# The key is the name of the node type, which is just for debugging purposes.
# The node config specifies the launch config and physical instance type.
available_node_types:
    cpu_4_ondemand:
        node_config:
            InstanceType: m4.xlarge
        # For AWS instances, autoscaler will automatically add the available
        # CPUs/GPUs/accelerator_type ({"CPU": 4} for m4.xlarge) in "resources".
        resources: {"CPU": 4}
        min_workers: 2
        max_workers: 5
    custom1:
        node_config:
            InstanceType: m4.4xlarge
            InstanceMarketOptions:
                MarketType: spot
        resources: {"Custom1": 1}
        max_workers: 10
    custom2:
        node_config:
            InstanceType: m4.4xlarge
            InstanceMarketOptions:
                MarketType: spot
        resources: {"Custom2": 2}
        max_workers: 4

# Specify the node type of the head node (as configured above).
head_node_type: cpu_4_ondemand

# Specify the default type of the worker node (as configured above).
worker_default_node_type: cpu_4_ondemand

# The default settings for the head node. This will be merged with the per-node
# type configs given above.
head_node:
    ImageId: latest_dlami

# The default settings for worker nodes. This will be merged with the per-node
# type configs given above.
worker_nodes:
    ImageId: latest_dlami

setup_commands:
    - pip install -U https://s3-us-west-2.amazonaws.com/ray-wheels/latest/ray-1.1.0.dev0-cp37-cp37m-manylinux2014_x86_64.whl
    - git clone --branch autoscaler_formatted_report https://github.com/wuisawesome/ray.git || true
    - pushd ray; git reset --hard; popd;
    - ray/python/ray/setup-dev.py -y 
    - rm /home/ubuntu/anaconda3/lib/python3.7/site-packages/ray/monitor.py
    - ln -s /home/ubuntu/ray/python/ray/monitor.py /home/ubuntu/anaconda3/lib/python3.7/site-packages/ray/monitor.py


# Configure the cluster for very conservative auto-scaling otherwise.
target_utilization_fraction: 1.0
idle_timeout_minutes: 2

# How Ray will authenticate with newly launched nodes.
auth:
    ssh_user: ubuntu