AWS Glue 6.0 Released with 30% Price Reduction and Full Apache Iceberg v3 Support

AWS announced the general availability of AWS Glue 6.0. This version offers 30% lower pricing compared to previous versions and improves performance with a modernized runtime based on Apache Spark 4.1, Python 3.12, and Scala 2.13.

AWS Glue 6.0 provides the most complete Apache Iceberg v3 implementation among serverless managed Spark services. In particular, support for the VARIANT data type and shredding significantly improves query read performance for semi-structured data. This allows for efficient processing of JSON and log data without pipeline interruptions during schema changes.

Key new features include:

• Geospatial support: Native support for Geometry/Geography data types for GIS analysis and location-based processing.
• Nanosecond precision timestamps: Enhanced precision for IoT and high-frequency financial workloads.
• Spark Declarative Pipelines: Introduction of declarative pipelines to simplify ETL authoring.
• Arrow-native Python UDFs: Improved PySpark performance by eliminating serialization overhead.
• Real-time streaming mode: Support for real-time event processing achieving single-digit millisecond latency.

It is immediately available via AWS CLI, SDKs, and Glue Studio using the --glue-version parameter without any changes to existing APIs. The service is available in all AWS regions, and crawler and ETL jobs are billed on a per-second basis.