Cloudera Data Engineering Runtime end of support
Learn about Cloudera Data Engineering (CDE) Runtime end of support for Spark.
The following table specifies the planned End of Support (EoS) policy schedule for Spark. All future dates are provided for planning purposes only and are subject to change. In each case, the projected EoS date can be considered to be the last day of the month specified in the CDE Runtime end of support information table.
| Public Cloud Runtime version | Data Lake support | End of support | Long-term support | Notes |
|---|---|---|---|---|
| Spark 2.4.8 | Up to 7.2.18 | September 2025 | Yes | Deprecated (will only receive bug fixes and security patches) |
| Spark 3.2.3 | Up to 7.2.18 | September 2025 | - | - |
| Spark 3.3.x | Up to 7.2.18 | September 2025 | - | - |
| Spark 3.5.x | Starting from 7.2.18 | Will follow Data Lake support policy | Yes | - |
Frequently Asked Questions (FAQs)
What is the relationship between CDE Spark runtimes and the Data Lake version?
Each CDE release is certified against one or more Data Lake versions. This ensures full compatibility between the Spark engine and the underlying platform capabilities, such as the Hive Catalog service and Ranger for security policies. Each Spark version is compatible with a Data Lake version. For more information, see Compatibility for Cloudera Data Engineering and Runtime components.
Does CDE offer Long-term support (LTS) releases?
CDE offers LTS through underlying Spark runtimes. When running Spark jobs within CDE, you have the option to choose an older Spark version. Specific versions of Spark are designated as LTS. This allows you to continue running Spark jobs without any code changes. Since CDE job management APIs remain backwards compatible, existing automations are not impacted.
What is the EoS timeline for Spark runtimes designated LTS?
Spark runtimes designated as LTS in the Public Cloud have certification that spans through multiple Data Lake versions and are limited to minor enhancements and critical bug and security enhancements. To use a Spark runtime that is tagged as LTS, you need to run a supported CDE version with a corresponding certified and currently supported Data Lake version.
What is the EOS timeline for non-LTS Spark runtimes?
Spark runtimes that are not designated as LTS will be limited to the support lifetime of the corresponding certified Data Lake version. For more information, see Compatibility for Cloudera Data Engineering and Runtime components.
What is the support timeline for a given Data Lake version?
See the Cloudera Data Platform - Public Cloud section in Support lifecycle policy.
