ADA-C01 Question 264
Single answerIdentify use casesA Snowflake administrator supports a business unit that loads customer transaction files from an internal stage every 15 minutes into a curated reporting table. The source files occasionally contain malformed rows, and analysts require the load process to continue so valid rows are available quickly while invalid rows can be reviewed later. The team also wants a simple way to identify which files have already been processed and avoid reloading them accidentally. Which Snowflake feature is the BEST fit for this use case?
- A
Use Snowpipe to continuously load files from the stage, with COPY options such as ON_ERROR=CONTINUE so valid rows load while rejected rows can be investigated separately
- B
Use a materialized view on top of the staged files so Snowflake can automatically ingest only valid records and skip invalid rows
- C
Use database replication so staged files are copied to another database, where invalid rows are filtered before loading into the reporting table
- D
Use a secure data share from the source system so analysts can query staged files directly without any loading process
Show answer and explanation
Correct answer: A
Explanation
This scenario is fundamentally about identifying the appropriate Snowflake use case and matching it to the right administrative feature. Snowpipe is the best fit because it is purpose-built for continuous, near-real-time ingestion of files from stages into Snowflake tables using COPY INTO semantics. For malformed records, COPY options such as ON_ERROR can be used so that valid rows continue loading while problematic rows are handled according to the configured error behavior. Snowflake also maintains load metadata and history that helps administrators determine which files were processed and avoid unintended duplicate loads. By contrast, materialized views are for query acceleration, replication is for object replication and disaster recovery/business continuity scenarios, and secure data sharing is for governed data access across accounts without ETL. These distinctions align with Snowflake documentation for Snowpipe, COPY INTO
, load history, replication, and secure data sharing best practices.- A. Correct.
Correct. Snowpipe is designed for continuous data loading from stages and is a strong fit when files arrive frequently and should be ingested with minimal latency. Snowpipe uses COPY semantics, including options such as ON_ERROR=CONTINUE, allowing valid rows to load even if some rows are malformed. Snowflake also tracks load history and file metadata to help prevent accidental reloading of the same files under normal loading patterns. This directly matches the operational need to load incremental files every 15 minutes while handling bad rows pragmatically.
- B. Incorrect.
Incorrect. Materialized views do not ingest files from stages. They are built on query results from tables or other supported relational objects and are used to precompute and maintain query results for performance. They do not provide staged file processing, row rejection handling during load, or file load tracking.
- C. Incorrect.
Incorrect. Database replication is intended for replicating supported Snowflake objects across regions or accounts for business continuity and related use cases. It is not a file ingestion mechanism and does not solve the requirement to continuously process new staged files, tolerate malformed input rows during load, or track which files have already been loaded into a target table.
- D. Incorrect.
Incorrect. Secure data sharing allows Snowflake data to be shared without copying data, but it does not expose staged files from an internal stage as a substitute for ingestion into reporting tables. It also does not address malformed-row handling during file loading or file-level load history management. A candidate might choose this if they confuse data sharing with data ingestion.