Glue 6.0 updates the engine behind data processing

A new execution foundation

On 21 August 2026, AWS announced general availability of Glue 6.0, with Spark 4.1, Python 3.13, Scala 2.13 and Apache Iceberg v3 support.

An update that affects data contracts

Glue prepares and transforms data. Iceberg organizes data-lake tables while files remain in object storage. For an analytics team, updating this foundation can concern formats consumed by other tools as much as processing speed.

A useful starting point would be a representative batch: a daily transformation, a custom Python function and a table shared with another engine. Comparing output rows, types, null values and processing times would separate compatibility from performance.

Engine pricing is only part of the bill

A lower unit rate cannot guarantee a smaller total bill if durations, volumes or retries change. Dependencies and table readability for existing consumers also need checking. A successful migration first preserves what the data means, then demonstrates a measured improvement across the complete job.

Official source.

Editorial retrospective: event dates are stated in the text.