Using Pinned EMR and Databricks Runtime Releases

Overview

By default, a Tecton materialization cluster uses a specific EMR release or Databricks Runtime release, which computes your feature values for online serving and training dataframes. Periodically, Tecton upgrades the default EMR release/Databricks Runtime release on materialization clusters, to apply the latest security patches and stability fixes. These upgrades may include Spark upgrades.

The following lists show the supported Databricks Runtime and EMR Versions and the defaults for Tecton 0.5 and 0.6:

Supported Databricks Runtimes:

• 9.1.x-scala2.12 (Tecton 0.5 default)
• 10.4.x-scala2.12 (Tecton 0.6 default)
• 11.3.x-scala2.12

Supported EMR Versions:

• emr-6.5.0 (Tecton 0.5 default)
• emr-6.7.0 (Tecton 0.6 default)
• emr-6.9.0

Rarely, existing transformation logic defined in Tecton will be incompatible with a Spark upgrade.

To prevent a Spark upgrade (that will occur due to a EMR upgrade/Databricks Runtime upgrade), or to downgrade Spark if an incompatibility has occurred, you can configure Tecton to override the default EMR release/Databricks Runtime release, per Feature View and Feature Table.

Overriding Tecton’s default EMR release/Databricks Runtime release

In Feature View and Feature Table definitions, you can specify which EMR release/Databricks Runtime release is used, by setting the parameters in the table below to a DatabricksClusterConfig or a EMRClusterConfig object.

Object	Parameter to Set
@batch_feature_view	batch_config
@stream_feature_view	stream_config
FeatureTable	batch_config

If using a DatabricksClusterConfig object, set the dbr_version parameter. Note: the name must be a valid runtime name.

For example:

@batch_feature_view(
    batch_compute=DatabricksClusterConfig(dbr_version="9.1.x-scala2.12"),
    # ...
)
def my_feature_view(input_data):
    pass

If using a EMRClusterConfig object, set the emr_version parameter.

For example:

@batch_feature_view(
    batch_compute=EMRClusterConfig(emr_version="emr-6.5.0"),
    # ...
)
def my_feature_view(input_data):
    pass

Upgrade Considerations

Transformation logic

Custom transformation logic defined in Tecton may have some library and version dependencies that are no longer supported in the new EMR / DBR runtime environments.

Please see How to manage Tecton versions and safely upgrade to help mitigate risk.

Delta + EMR + Spark Upgrade

For EMR users of the Delta offline Store (offline_store=DeltaConfig()):

1. When upgrading from emr-6.5.0 (Tecton 0.5 default), verify that no concurrent jobs are running on emr-6.5.0 and emr-6.7.0+ for the same feature view. To avoid this, pause materialization before updating the emr version to guarantee a smooth transition.
2. The newer delta-core libraries rely on dynamodb:GetItem action to get the previous delta transaction. Please verify that this action is allowed in your EMR policy before upgrading.

Looking at the Spark release notes

We recommend looking at the Spark release notes to see if your Tecton transformations are using any deprecated features, and check if any custom JARs you use need to be updated to be compatible. This page contains links to the release notes for each Spark version.

The links below show the Spark version that is included in each version of Databricks Runtime and EMR, respectively:

• Databricks runtime releases

• Spark release history (EMR)