Spark, Hive, and Presto-compatible jobs read the same S3 tables.
Share table and partition metadata instead of maintaining separate local DDL per cluster.
Which configuration is most appropriate?
Use Glue Data Catalog as a Hive-compatible metastore for EMR engines.
Detailed explanation
Local definitions diverge and make schema and partition updates inconsistent.
Local definitions diverge and make schema and partition updates inconsistent.
Object protection does not provide metastore query and update APIs.
Object protection does not provide metastore query and update APIs.
Metrics contain numeric observations, not table columns and partitions.
Metrics contain numeric observations, not table columns and partitions.
EMR engines can share table and partition definitions through the managed catalog.
EMR engines can share table and partition definitions through the managed catalog.
Check official documentation
Compare the following points with the official documentation.
EMRのGlue Data Catalog連携、Hive metastore、Lake Formation権限の公式説明を確認する。
What to confirm
ローカルDDL・オブジェクト保護・メトリクスと共有メタストアの違いを説明できる。
Key points
- Glue Data Catalog
- Hive metastore
- EMR
- Shared metadata
Notes
- Environment: AWS公式DEA-C01試験ガイドと各サービスの公式ドキュメントの確認
- Confirm current AWS service behavior in the official DEA-C01 documentation.
Foundation review
Shared metadata
A central metastore prevents local DDL copies from diverging.
Governed access
Catalog sharing still requires IAM, S3, and Lake Formation permissions.