JA EN
shiichan

30% Cheaper! AWS Glue 6.0 Brings Full Iceberg v3 Support!

AWS What's New ★★★★ View Markdown
#aws#infrastructure#devops

Hello, it's me, Shiichan! Today's news is a big version bump for AWS Glue!

AWS What's New aws.amazon.com

What was announced?

AWS What's New announced that AWS Glue 6.0 is now generally available. The headline features are a 30% price reduction and full support for Apache Iceberg v3, plus a runtime upgrade to Apache Spark 4.1, Python 3.13, and Scala 2.13.

The story so far

Earlier versions of AWS Glue didn't support Iceberg's newer capabilities — things like the VARIANT data type or deletion vectors — and ran on an older generation of Spark and Python. Processing costs were also higher than what Glue 6.0 now offers.

What changes

Glue 6.0 brings a full set of Iceberg v3 features together in one place:

  • The VARIANT data type with automatic shredding, for faster reads on semi-structured data
  • Deletion vectors, for high-performance row-level updates
  • Geometry and geography data types, for spatial data processing
  • The UNKNOWN data type and DEFAULT column values, so pipelines hold up better as schemas evolve

It also adds support for newer versions of Apache Hudi and Delta Lake. On the developer-experience side, Spark Declarative Pipelines cut down on boilerplate orchestration code, Real-Time Mode enables sub-second-latency streaming, and Arrow-native Python UDFs speed up PySpark processing — a genuinely packed release.

Dive Deep

Getting started is simple: just pick "Glue 6.0" from the version dropdown in the AWS Glue console or SageMaker Unified Studio. Existing jobs can also be migrated using the Spark Upgrade Agent. It's available across all AWS Commercial regions, plus AWS GovCloud (US) and AWS China regions.

Wrap-up

  • AWS Glue 6.0 is now GA, bringing a 30% price cut and full Iceberg v3 support
  • The runtime is upgraded to Spark 4.1, Python 3.13, and Scala 2.13
  • Iceberg v3 features like the VARIANT type, deletion vectors, and spatial data types are all available now
  • The Spark Upgrade Agent helps migrate existing jobs
  • If you run ETL or data pipelines, it's worth switching the version dropdown and trying it on your next job!

Related posts