November 19, 2026 · Google CloudAnnouncement
Managed Service for Apache Spark image version 4.0 generally available
Managed Service for Apache Spark (formerly Dataproc on Compute Engine): cluster image version 4.0 is generally available (GA) with the following image versions:
4.0.0-debian13,4.0.0-ml-ubuntu24,4.0.0-rocky9, and4.0.0-ubuntu24- The
4.0.0-ml-ubuntu24image extends the4.0base image with machine learning libraries and GPU support.
- The
Image version 4.0 is a lightweight image that contains the following
pre-installed components, reducing exposure to Common Vulnerabilities and
Exposures (CVEs):
- Apache Gluten 1.6
- Apache Hadoop 3.5.0
- Apache Hive 4.2.0
- Apache Iceberg 1.12.0 (pre-installed optional component)
- Apache Spark 4.2.0
- Apache Tez 0.10.5
- Cloud Storage Connector 4.0.7
- Conscrypt 2.6
- Delta Lake 4.4.0 (pre-installed optional component)
- Java 25 for Spark workloads, with Java 21 for backward compatibility
- Python 3.13 with Pixi 0.76.0
- R 4.5
- Scala 2.13.17
- Spark-BigQuery Connector 0.46.0
You can activate the pre-installed Apache Iceberg and Delta Lake components, or
install additional
optional components,
when you create a 4.0 cluster.
Recommendation: Use image version 2.3 or later when creating a Managed
Service for Apache Spark cluster to meet security compliance requirements.
Text from the Google Cloud release notes, licensed under CC BY 4.0. Source: Google Cloud release notes.
Free. Unsubscribe any time.
Related updates
Managed Service for Apache Spark runtime 4.0 GA, with Spark 4.2.0 and Java 25
Managed Service for Apache Spark adds image versions 2.2.88, 2.3.37 and 3.0.3
Managed Service for Apache Spark adds subminor cluster image 2.1.120
Managed Service for Apache Spark adds Serverless runtimes 1.2.90, 2.2.90, 2.3.43