marquez
marquez copied to clipboard
Have a strategy to deal with renamed datasets.
Some use cases to consider in this one include
- Catalogs that point to external datasets
- Hive tables that point to, e.g., S3 files
- Spark catalog tables that point to JDBC queries
- BigQuery tables that point to GCS files
- Delta mounts on Databricks clusters
- Postgres views
- Renaming datasets for organization purposes
- Friendly names for datasets instead of the file path
- Datasets namespaced by domain instead of physical location
@julienledem @collado-mike with the introduction of symlinks via https://github.com/MarquezProject/marquez/pull/2087, can we close this issue?