Tune the cluster¶

This section describes how to tune the DataRobot application by modifying ConfigMaps, specifically the datarobot-modeling-envvars ConfigMap, through Helm chart values.

Use ConfigMaps¶

DataRobot uses ConfigMaps to populate environment variables in cluster pods. Those variables are then read by services to determine their settings.

DataRobot Helm charts create and use many ConfigMaps. As a cluster administrator, you can configure or tune the services by modifying the datarobot-modeling-envvars ConfigMap.

To apply changes to the ConfigMap, you must provide the values in a YAML file and then perform a helm upgrade on the release.

Apply datarobot-modeling-envvars ConfigMap¶

The datarobot-modeling-envvars ConfigMap is controlled in the DataRobot umbrella charts with the core.config_env_vars setting. When Helm processes the release and its charts, the settings are placed into the datarobot-modeling-envvars.data section of the ConfigMap.

This method is very similar to the way the app_configuration.drenv_override section of the config.yaml file worked in DataRobot releases prior to 9.0.

To add a configuration option, create or add the core.config_env_vars setting in the values.yaml file used to apply the relevant umbrella chart. The example below is for the datarobot-generic chart with MinIO object storage:

core:
  config_env_vars:
    S3_HOST: core-minio.DR_CORE_NAMESPACE.svc.cluster.local
    LOGGING_LEVEL: INFO
    LOG_LEVEL: INFO
    AWS_ACCESS_KEY_ID: miniodatarobot
    AWS_SECRET_ACCESS_KEY: miniodatarobot 

For example, to adjust the number of Model Management workers, you can override the default value of two (2) workers. To increase the count to four (4), add the MMM_MODMON_WORKER_COUNT key and the value "4" to your configuration. The updated section would look like this:

core:
  config_env_vars:
    S3_HOST: core-minio.DR_CORE_NAMESPACE.svc.cluster.local
    LOGGING_LEVEL: INFO
    LOG_LEVEL: INFO
    AWS_ACCESS_KEY_ID: miniodatarobot
    AWS_SECRET_ACCESS_KEY: miniodatarobot
    MMM_MODMON_WORKER_COUNT: "4" 

To apply the change to the cluster, run the helm upgrade command for your release.