Databricks Runtime 9.1 LTS migration guide

This guide helps you migrate your Databricks workloads to Databricks Runtime 9.1 LTS, powered by Apache Spark 3.1.2.

Databricks recommends that you migrate your workloads to a supported LTS version from that version’s next most-recent supported LTS version whenever possible. Therefore, this article covers Databricks Runtime versions from Databricks Runtime 7.3 LTS to Databricks Runtime 9.1 LTS.

Note

LTS means this version is under long-term support. See Long-term support (LTS) lifecycle.

To migrate workloads from Databricks Runtime 7.2 or below, Databricks recommends that you migrate your workloads in the following order:

  1. Migrate to Databricks Runtime LTS 7.3. See the Databricks Runtime 7.3 LTS migration guide.

  2. Migrate from Databricks Runtime 7.3 LTS to Databricks Runtime 9.1 LTS by following the Apache Spark migration guidance.

Databricks Runtime 11.3 LTS is the latest supported LTS version. To migrate workloads from Databricks Runtime 9.1 LTS to Databricks Runtime 11.3 LTS, see the Databricks Runtime 11.x migration guide.

Apache Spark migration guidance

Databricks Runtime 9.1 LTS is powered by Apache Spark 3.1.2. For Spark-specific migration information, see the Apache Spark 3.1.2 Migration Guide. For reference, the following Databricks Runtime versions from Databricks Runtime 7.3 LTS to Databricks Runtime 9.1 LTS are powered by these Apache Spark versions:

Databricks Runtime version

Apache Spark version

8.4 - 9.1 LTS

3.1.2

8.0 - 8.3

3.1.1

7.5 - 7.6

3.0.1

7.4

3.0.0

7.3 LTS

3.0.1