Amazon S3 Tables now support all Apache Iceberg V3 data types

Amazon S3 Tables now support all Apache Iceberg V3 data types and features, enabling more efficient analytics on large datasets.

• Deletion vectors replace positional delete files with compact binary format, reducing compaction time and overhead
• Row lineage adds _row_id and _last_updated_sequence_number for tracking changed rows without full table scans
• New data types include variant for semi-structured data, nanosecond timestamps, geometry/geography for geospatial data, and unknown type
• Variant data is shredded into hidden columns with statistics enabling file pruning and reduced I/O
• Upgrade existing V2 tables atomically without rewriting data using ALTER TABLE command
• V3 data types require Apache Spark 4.0 or later (AWS Glue 6.0+, Amazon EMR 8.1+)
• Available in all AWS Regions where S3 Tables are supported at no additional charge

S3 Tables fully manages V3 compaction and maintenance, enabling teams to handle compliance deletes, semi-structured data, and geospatial analytics more efficiently.