Amazon EMR Introduces Long Term Support for Apache Spark with Version 4.1
Amazon EMR now offers Long Term Support (LTS) releases for Apache Spark, beginning with emr-spark-8.1.0 and Spark 4.1, providing 36 months of support for critical security, bug, and data-corruption fixes. This allows users to run production Spark workloads on a single release for an extended period, facilitating more flexible upgrade schedules at no extra cost. The release also brings full support for Apache Iceberg v3, enhancing geospatial, high-precision timestamp, and schema evolution capabilities, alongside improved Spark SQL query features. Fine-grained access control has been extended, and EMR on EKS clusters now support Spark Connect endpoints with token-based authentication, all generally available across EMR deployment options.
- →Long Term Support (LTS) for Amazon EMR Apache Spark
- →Full Apache Iceberg v3 Support
- →Enhanced Spark SQL Catalog References and Auto-Detection
- →Spark Connect Endpoints for EMR on EKS
- →Expanded Fine-Grained Access Control
Features (4) ›
- Long Term Support (LTS) for Amazon EMR Apache Spark
Amazon EMR introduces LTS releases, starting with emr-spark-8.1.0 and Apache Spark 4.1, offering 36 months of support. This includes fixes for critical and high severity security, bug, and data-corruption issues, helping users run production Spark workloads longer and manage upgrades on their own schedule.
- Full Apache Iceberg v3 Support
The emr-spark-8.1.0 release adds full support for Apache Iceberg v3. This brings new geospatial, high-precision timestamp, and schema-evolution capabilities to tables used with Amazon EMR.
- Enhanced Spark SQL Catalog References and Auto-Detection
Spark SQL queries can now reference catalogs by name, including cross-account and Amazon S3 Tables catalogs. The system automatically detects Apache Iceberg, Delta Lake, and Apache Hudi table formats without requiring manual registration in Spark configuration.
- Spark Connect Endpoints for EMR on EKS
Amazon EMR on EKS clusters now support Spark Connect endpoints. These endpoints come with token-based authentication, enhancing connectivity and security for applications running on EKS.
Enhancements (1) ›
- Expanded Fine-Grained Access Control
Fine-grained access control has been extended to cover more Apache Iceberg operations and the Delta Lake VACUUM operation. This allows for the application of column-level and row-level permissions to a broader range of jobs.
https://aws.amazon.com/about-aws/whats-new/2026/09/amazon-emr-long-term-support-spark-4-1/
Related releases
- Amazon S3 Express One Zone Expands to 7 New AWS Regions AWS What's New ·
- AWS HealthOmics Adds IAM Session Policies for Dynamic Run Permissions AWS What's New ·
- Amazon ECS extends S3 Files support to EC2 launch type for containerized applications AWS What's New ·
- AWS Glue zero-ETL Integrations Gain Table Property Ownership and Conflict Detection AWS What's New ·
- AWS Lambda adds direct read configuration for Amazon S3 Files AWS What's New ·
- SageMaker HyperPod adds model caching for faster inference autoscaling AWS What's New ·