ADA-C01 Question 184
Single answerIdentify tagging use casesA Snowflake administrator is helping a healthcare company improve data governance across hundreds of tables in multiple schemas. The company wants to classify columns that contain sensitive data such as Social Security numbers, email addresses, and patient identifiers so that governance teams can search for those objects later and apply masking policies consistently. Which Snowflake feature is the best fit for this requirement?
- A
Create tags such as DATA_CLASSIFICATION and apply them to sensitive columns
- B
Create resource monitors for each schema and use them to identify sensitive columns
- C
Use clustering keys on sensitive tables so governance teams can find protected data more easily
- D
Create streams on each table to track which columns contain regulated data
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use tags. In Snowflake, tags are intended for assigning custom metadata to supported objects, including columns, making them well suited for data governance, classification, cost attribution, and ownership tracking use cases. For this scenario, the key need is to identify sensitive columns across many objects and later use that metadata operationally. Tags support exactly that pattern. They also integrate with governance workflows, including tag-based masking policy strategies in Snowflake. By contrast, resource monitors address spend control, clustering keys address performance optimization, and streams address CDC. Snowflake documentation on object tagging and data governance describes tags as a core feature for classifying and organizing data assets.
- A. Correct.
Correct. Tags are designed to attach business or governance metadata to Snowflake objects, including columns. A tag like DATA_CLASSIFICATION can store values such as PII, PHI, or CONFIDENTIAL, allowing administrators and governance teams to identify sensitive data across databases and schemas. Tags are a common mechanism for organizing objects for governance use cases and can also be used with tag-based masking policies to help enforce consistent protection.
- B. Incorrect.
Incorrect. Resource monitors are used to control and monitor credit consumption for virtual warehouses and, at the account level, cloud services. They do not classify data or annotate columns with governance metadata. This distractor reflects a misunderstanding between cost governance and data governance features.
- C. Incorrect.
Incorrect. Clustering keys improve pruning and query performance on large tables with selective predicates, but they do not provide metadata for identifying or classifying sensitive columns. Someone might choose this option if they confuse physical optimization features with governance and discovery features.
- D. Incorrect.
Incorrect. Streams capture change data capture (CDC) information for tables, views, or external tables so downstream processes can consume inserts, updates, and deletes. They do not provide a mechanism for labeling columns as sensitive or for storing classification metadata.