Which of the following describes how clustering keys work in Snowflake?

Question

Options

• AClustering keys update the micro-partitions in place with a full sort, and impact the DML
• BClustering keys sort the designated columns over time, without blocking DML operations
• CClustering keys create a distributed, parallel data structure of pointers to a table's rows and
• DClustering keys establish a hashed key on each node of a virtual warehouse to optimize joins at

• A
  3% (1)
• B
  86% (31)
• C
  8% (3)
• D
  3% (1)

Explanation

Snowflake clustering keys work through an asynchronous background process called Automatic Clustering. Over time, Snowflake reorganizes micro-partitions so that rows sharing similar clustering key values are co-located, improving pruning efficiency for range-based queries. Critically, this process runs in the background and does not block or impact concurrent DML operations (inserts, updates, deletes). Option A is wrong because micro-partitions are immutable - they are never updated in place; new micro-partitions are written and old ones retired. Option C describes an index structure, which Snowflake does not use. Option D describes hash-based join optimization at the warehouse level, which is unrelated to clustering.

No community discussion yet for this question.