Seeing a lot of excitement for row-level security (which is amazing, no doubt!) but my team is more excited about column-level security. Being in banking, a lot of records are peppered with PII and the governance over who gets to see what (combination of) columns can get messy. With CLS, we can implement fact and one-big tables without having to resort to view gymnastics (i.e. using views to hash columns depending on the viewer) or maintaining multiple pipelines for the same table with different audiences. Just another win for Unity Catalog and Databricks.
Thrilled to announce the GA of Row and Column Level Security in Unity Catalog! For the first time on any platform, you can apply fine-grained access controls across Apache Spark, data warehousing, AI, and real-time processing—all within a single platform. Unlock seamless, secure data collaboration without the complexity! https://lnkd.in/gFTgCrQY
