Runtime settings for Docker, resources, and stack components

Using settings to configure runtime configuration.

Stack Component Config vs Settings in ZenML

Part of the configuration of a pipeline are its Settings. These allow you to configure runtime configurations for stack components and pipelines. Concretely, they allow you to configure:

• The resources required for a step

• Configuring the containerization process of a pipeline (e.g. What requirements get installed in the Docker image)

• Stack component-specific configuration, e.g., if you have an experiment tracker passing in the name of the experiment at runtime

You will learn about all of the above in more detail later, but for now, let's try to understand that all of this configuration flows through one central concept called BaseSettings. (From here on, we use settings and BaseSettings as analogous in this guide).

Types of settings

Settings are categorized into two types:

• General settings that can be used on all ZenML pipelines. Examples of these are:

  • DockerSettings to specify Docker settings.

  • ResourceSettings to specify resource settings.

• Stack-component-specific settings: These can be used to supply runtime configurations to certain stack components (key= <COMPONENT_CATEGORY>.<COMPONENT_FLAVOR>). Settings for components not in the active stack will be ignored. Examples of these are:

  • SkypilotAWSOrchestratorSettings to specify Skypilot settings (works for SkypilotGCPOrchestratorSettings and SkypilotAzureOrchestratorSettings as well).

  • KubeflowOrchestratorSettings to specify Kubeflow settings.

  • MLflowExperimentTrackerSettings to specify MLflow settings.

  • WandbExperimentTrackerSettings to specify W&B settings.

  • WhylogsDataValidatorSettings to specify Whylogs settings.

  • SagemakerStepOperatorSettings to specify AWS Sagemaker step operator settings.

  • VertexStepOperatorSettings to specify GCP Vertex step operator settings.

  • AzureMLStepOperatorSettings to specify AzureML step operator settings.

Difference between stack component settings at registration-time vs real-time

For stack-component-specific settings, you might be wondering what the difference is between these and the configuration passed in while doing zenml stack-component register <NAME> --config1=configvalue --config2=configvalue, etc. The answer is that the configuration passed in at registration time is static and fixed throughout all pipeline runs, while the settings can change.

A good example of this is the MLflow Experiment Tracker, where configuration which remains static such as the tracking_url is sent through at registration time, while runtime configuration such as the experiment_name (which might change every pipeline run) is sent through as runtime settings.

Even though settings can be overridden at runtime, you can also specify default values for settings while configuring a stack component. For example, you could set a default value for the nested setting of your MLflow experiment tracker: zenml experiment-tracker register <NAME> --flavor=mlflow --nested=True

This means that all pipelines that run using this experiment tracker use nested MLflow runs unless overridden by specifying settings for the pipeline at runtime.

Using the right key for Stack-component-specific settings

When specifying stack-component-specific settings, a key needs to be passed. This key should always correspond to the pattern: <COMPONENT_CATEGORY>.<COMPONENT_FLAVOR>

For example, the SagemakerStepOperator supports passing in estimator_args. The way to specify this would be to use the key step_operator.sagemaker

@step(step_operator="nameofstepoperator", settings= {"step_operator.sagemaker": {"estimator_args": {"instance_type": "m7g.medium"}}})
def my_step():
  ...

# Using the class
@step(step_operator="nameofstepoperator", settings= {"step_operator.sagemaker": SagemakerStepOperatorSettings(instance_type="m7g.medium")})
def my_step():
  ...

or in YAML:

steps:
  my_step:
    step_operator: "nameofstepoperator"
    settings:
      step_operator.sagemaker:
        estimator_args:
          instance_type: m7g.medium