Databricks Runtime 13.x migration guide
This guide helps you migrate your Azure Databricks workloads to the latest version of Databricks Runtime 13.x. To do so, Databricks recommends that you migrate your workloads in the following order:
- If your workloads are currently running on Databricks Runtime 12.2 LTS or above, you can migrate directly to the latest version of Databricks Runtime 13.x, as described later in this article.
- If your workloads are currently running on Databricks Runtime 12.2 LTS or below, do the following:
- Migrate to Databricks Runtime 12.2 LTS first. See the Databricks Runtime 12.x migration guide.
- Follow the guidance in this article to migrate from Databricks Runtime 12.2 LTS to the latest version of Databricks Runtime 13.x.
Note
LTS means this version is under long-term support. See Databricks Runtime LTS version lifecycle.
Apache Spark migration guidance
The most recent version of the Databricks Runtime is 13.3 LTS, powered by Apache Spark 3.4.1. For Spark-specific migration information, click on one or more of the following links by Databricks Runtime version to view the corresponding Apache Spark Migration Guide. For reference, Databricks Runtime 12.2 LTS is powered by Apache Spark 3.3.2.
Databricks Runtime version | Apache Spark version |
---|---|
13.3 LTS | 3.4.1 |
13.2 | 3.4.0 |
Databricks Runtime behavior changes
The following notable behavior changes were introduced into each of the following DBR 13.x releases.
Databricks Runtime system environment properties, features, and libraries
For information about system environment properties as well as new, changed, and deprecated features and libraries in Databricks Runtime releases from Databricks Runtime 12.2 LTS to the latest version of Databricks Runtime 13.x, see the following:
- Databricks Runtime 13.3 LTS for Machine Learning
- Databricks Runtime 13.3 LTS
- Databricks Runtime 13.2 for Machine Learning (EoS)
- Databricks Runtime 13.2 (EoS)
- Databricks Runtime 12.2 LTS for Machine Learning
- Databricks Runtime 12.2 LTS
Post-release maintenance updates are listed in Databricks Runtime maintenance updates.