Module 16 · Core

Catalogs: the source of truth

What a catalog does, why it holds the commit pointer, and the landscape from Hive Metastore to REST catalogs.

In plain words

A catalog is the lakehouse's directory: it knows every table, where its metadata lives and who may use it. Because every engine asks the same catalog, Spark, Trino and others all see the same tables and the same latest version.

Helps to take first

What makes a table a table

Words you'll meet

Catalog Metadata Query engine Commit

This module is in production

Here's what it will cover.

Centrepiece

Point Spark, Trino and DuckDB at one catalog and watch a commit from one become visible to the others

You will understand

Namespaces, table pointers and atomic commits
Hive Metastore, AWS Glue, Unity Catalog, Apache Polaris, Nessie
The Iceberg REST catalog spec
Git-like branching of data (Nessie, Iceberg branches)

Formats

Step-throughBuild & connectCheckpoints