Dataproc Serverless runtime versions

• Dataproc Serverless LTS (Long-Time-Support) runtime versions are supported for 30 months. Dataproc Serverless non-LTS runtime versions are supported for 12 months.
• Dataproc Serverless runtime versions continue to be available for two years after their end-of-support date.

Supported Dataproc Serverless versions

Version	Last Updated	Released On	Supported Until	Available Until	Notes
Spark runtime 2.1	2023/09/22	2023/05/05	2024/05/31	2026/05/31	General Availability release.
Spark runtime 2.0	2023/09/22	2022/10/21	2023/10/31	2025/10/31	General Availability release. This is the default version.
Spark runtime 1.1 LTS	2023/09/22	2023/01/27	2025/07/31	2027/07/31	General Availability release

How to choose a Dataproc Serverless runtime version

The current default Dataproc Serverless runtime version is listed in Supported Dataproc Serverless versions. You can use the Google Cloud console, gcloud CLI, and the Dataproc API to choose a different runtime version when you submit a batch workload.

Console

1. Open the Dataproc Create batch page in the Google Cloud console.
2. Under Container, in the Runtime version section, select one of the listed Dataproc Serverless runtime versions.

gcloud

Use the dataproc batches submit --version flag to specify a Dataproc Serverless runtime version.

Example:

gcloud dataproc batches submit JOB_TYPE \
    --region=REGION \
    --version=VERSION (for example, 2.1)
    other args ...

API

Use the RuntimeConfig.version field as part of a batches.create API call to specify a Dataproc Serverless runtime version.

Unsupported Dataproc Serverless versions

The following Dataproc Serverless versions are unsupported.

Version	Includes	Released On	Last Updated	Available Until	Notes
Spark runtime 1.0	Apache Spark 3.2.3 Cloud Storage connector 2.2.11 Conda 4.11 Java 11 Python 3.9 R 4.1 Scala 2.12.17	2022/01/19	2023/2/03	2025/01/31	Unsupported as of 2023/02/03. 1.0.29 was the final released version.