The data engineering team is migrating an enterprise system with thousands of tables and views into the Lakehouse. They plan to implement the target architecture using a series of bronze, silver, and gold tables. Bronze tables will almost exclusively be used by production data engineering workloads, while silver tables will be used to support both data engineering and machine learning workloads. Gold tables will largely serve business intelligence and reporting purposes. While personal identifying information (PII) exists in all tiers of data, pseudonymization and anonymization rules are in place for all data at the silver and gold levels. The organization is interested in reducing security concerns while maximizing the ability to collaborate across diverse teams. Which statement exemplifies best practices for implementing this system?
Select one answer.
Marked answer: A
Community Discussion
Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.
Chose A1
suretly A
Chose A3
A - most logical B - it is a logical construct, but under default settings tables are stored where the database is so there is a security component to it C - never a good idea to store everything in one db since db allows to group tables with similar area of interest and allows to manage permissions (like groups in Entra and assigning permissions to groups) D - not default database does not mean we cannot use managed tables and you can specify your location still; I do not think that storing anything on DBFS is a good idea - even Databricks suggests to use workspaces for your code, not to mention the data. E - thousand databases - nonsense; you can specify the location of individual tables.
1
Correction to D explanation - apparently the default location is bfs
