Ravi Sir's Cards › AWS
Glue Data Catalog
This revision card explains Glue Data Catalog. It starts from the need that without a catalog every tool reinvents schema and breaks on drift. The catalog is a Hive-compatible metastore of databases/tables/partitions pointing at S3. The how-to section covers how to use crawlers or IaC to register tables; keep schema evolution intentional. An example on the card is athena, Redshift Spectrum, and EMR all see the same sales table definition.
Read card text
What this card says
Why do we need it?
Without a catalog every tool reinvents schema and breaks on drift.
What is it?
The catalog is a Hive-compatible metastore of databases/tables/partitions pointing at S3.
catalog
How to do it
Use crawlers or IaC to register tables; keep schema evolution intentional.
Example
Athena, Redshift Spectrum, and EMR all see the same sales table definition.