Table formats such as Iceberg, Hudi and Delta build on this and add extra capabilities:

A table like structure and schema, meaning that we can treat files as tables with rows and columns;

Inserts, updates, deletes and merge operations meaning that we can update them as we would a relational database table;

Transactional safety, meaning that multiple users can be inserting, updating and reading from the tables concurrently without seeing inconsistent data;

By using these table formats, we take a big step towards turning data lakes into databases by making them well structured, easier to work with and scalable to more concurrent users.

[https]
Open Table Formats Are Inevitable For Analytical Datasets | Ensemble

--
[Cookie]
[Received]
[Client]