Apache Iceberg is an open table format for very large analytic tables on data lakes. It brings the reliability and simplicity of SQL tables to big data: you can merge, update and selectively delete rows, add, rename or reorder columns without rewriting the table, and use time travel to read or roll back to earlier versions. Iceberg is a specification with libraries (including for Java, Python, Rust and Go), not a hosted service.
Iceberg started at Netflix in 2017, was donated to the Apache Software Foundation, a US non-profit organisation, in late 2018, and has been a top-level project since May 2020 under the Apache License 2.0. You run Iceberg yourself on storage of your choice, in your own data centre or with a European provider, so you decide where your data resides. The open format limits lock-in and makes it possible to switch query engine or supplier.