## Configure the task container defaults. Tasks include trials, commands, tensorboards and more.
## For all task containers, shm_size_bytes and network_mode are configurable. For trials, the
## network interface used by distributed (multi-machine) training is configurable.
#task_container_defaults:
#  shm_size_bytes: 4294967296
#  network_mode: bridge
#  dtrain_network_interface: <network interface name>

## Resource manager configuration. Defaults to using the agent resource manager.
#resource_manager:
#  type: agent
#  scheduler:
#    type: fair_share
#    fitting_policy: best
#  default_cpu_resource_pool: default
#  default_gpu_resource_pool: default
## Resource pools configuration.
#resource_pools:
#  - pool_name: default

## The maximum CPU containers for the agent. Defaults to 100.
#    max_cpu_containers_per_agent: 100

## The provider of dynamic agents.
#    provider:

## The url of the Determined master.
#      master_url: <scheme://host:port>

## The startup script for the agent. This runs on the node the agent runs on.
#      startup_script: <startup script>

## The startup script for the agent's container. This runs in the container determined-agent runs in.
#      container_startup_script: <container startup script>

## The Docker network to use for the agent when using dynamic agents. If this is
## set to "host", Docker host-mode networking will be used instead. Defaults to "default".
#      agent_docker_network: default

## The docker runtime to use for the agent when using dynamic agent. Defaults to "runc.
#      agent_docker_runtime: runc

## The docker image to use for the agent when using dynamic agents. This value
## must be configured.
#      agent_docker_image: determinedai/hpe-mlde-agent:<version>

## The maximum idle period of agents. The master waits for this period of time
## before shutting down idle agents. Defaults to 5 min.
#      max_idle_agent_period: 5m

## The maximum starting period of agents. The master waits for this period of time
## for starting agents before retrying. Defaults to 5 min.
#      max_agent_starting_period: 5m

## Configure AWS dynamic agents. The `provider`, `image_id`, `security_group`, and
## `ssh_key_name` are required to be set to use AWS dynamic agents.
#      type: aws
#      region: <region>
#      root_volume_size: 200
#      image_id: <AMI id>
#      tag_key: <tag key for agent discovery>
#      tag_value: <tag value for agent discovery>
#      instance_name: determined-ai-agent
#      ssh_key_name: <ssh key name>
#      iam_instance_profile_arn: <iam_instance_profile_arn>
#      network_interface:
#        public_ip: true
#        security_group_id: <security group id>
#        subnet_id: <subnet id>
#      instance_type: g4dn.metal
#      max_instances: 5

## Configure GCP dynamic agents. The `provider`, `boot_disk_source_image`,
## `network_interface` are required to be set to use GCP dynamic agents.
#      type: gcp
#      base_config: <instance resource base configuration>
#      project: <project id>
#      zone: <zone>
#      boot_disk_size: 200
#      boot_disk_source_image: projects/<project-id>/global/images/<image-name>
#      labels: {<key1>: <val1>, <key2>: <val2>, ...}
#      label_key: <label key for agent discovery>
#      label_value: <label value for agent discovery>
#      name_prefix: <name prefix>
#      network_interface:
#        network: projects/<project>/global/networks/<network>
#        subnetwork: projects/<project>/regions/<region>/subnetworks/<subnetwork>
#        external_ip: false
#      network_tags: ["<tag1>", "<tag2>"]
#      service_account:
#        email: "<service account email>"
#        scopes: ["https://www.googleapis.com/auth/cloud-platform"]
#      instance_type:
#        machine_type: n1-standard-32
#        gpu_type: nvidia-tesla-t4
#        gpu_num: 4
#      max_instances: 5
#      operation_timeout_period: 5m

## Configure default checkpoint storage configuration.
checkpoint_storage:
  type: shared_fs
  host_path: /tmp
  storage_path: determined-checkpoint
  save_experiment_best: 0
  save_trial_best: 1
  save_trial_latest: 1

## Configure the location where the master looks for static files.
# root: /usr/share/determined/master

## Configure the connection to the Postgres database.
db:
  user: postgres
  # host: determined-db
  port: 5432
  name: determined
  # password: database_password

observability:
  enable_prometheus: false

## A user friendly name to identify this cluster by.
# cluster_name: Dev

## Configure whether we collect anonymous information about the usage of
## Determined.
# telemetry:
#   enabled: true

## Configure the credentials for using the SCIM server to provision users.
# scim:
#   enabled: false
#   auth:
#     # "oauth" can be used instead of "basic"; then there should be no
#     # username or password fields.
#     type: basic
#     username: scim_username
#     password: scim_password

## Configure the settings for using the SAML server to sign in.
# saml:
#   enabled: false
#   provider: okta
#   idp_sso_url: sso_url
#   idp_sso_descriptor_url: descriptor_url
#   idp_recipient_url: recipient_url
#   idp_cert_path: cert_path