A team wants a table that automatically stays up to date from a defined query, with Snowflake managing the refresh to a freshness target, instead of building their own stream-and-task pipeline. What should they use?
- AA Dynamic Table with a TARGET_LAG.
- BA materialized view with manual refresh.
- CA standard view refreshed by a task.
- DA temporary table rebuilt each session.
Dynamic Tables are declarative: you define the query and a TARGET_LAG (freshness target), and Snowflake automatically refreshes them and orders dependencies, replacing many manual stream-and-task pipelines.
Why each option is right or wrong.
- A (correct). A Dynamic Table is defined by a query and auto-refreshes to the specified target lag, with Snowflake managing dependencies.
- B. Materialized views auto-maintain but have narrower query support and no target-lag freshness model for pipelines.
- C. A standard view stores no data and would recompute on every read, not maintain a refreshed table.
- D. A temporary table drops at session end and is not automatically refreshed.
Reference: SnowPro Core COF-C03, Domain 4 - Data Loading, Unloading, and Connectivity.