A Data Engineer is developing a data pipeline to ingest data from an Amazon S3 bucket into a Snowflake table. Files will be loaded into the S3 bucket randomly, with most files loaded during working ho

Question

Options

• ASnowpipe
• BSnowpipe Streaming
• CSnowflake Kafka Connector
• DThe COPY INTO <location> command

• A
  81% (25)
• B
  6% (2)
• C
  3% (1)
• D
  10% (3)

Explanation

Snowpipe is the correct choice because it is Snowflake's serverless, event-driven file ingestion service that integrates natively with S3 event notifications - no infrastructure to manage, you pay only for what you load, and files are ingested within minutes of arrival, easily satisfying the 10-minute SLA.

B (Snowpipe Streaming) is designed for high-frequency, row-level streaming via the Snowflake SDK, not file-based S3 loads - it carries higher cost and complexity than needed for 100 MB CSV files arriving intermittently.

C (Kafka Connector) requires provisioning and maintaining a Kafka cluster, introducing substantial operational overhead and cost that directly violates the top two priorities.

D (COPY INTO <location>) is actually a data unload command (Snowflake → external stage), so it's the wrong direction entirely; even if the intent was COPY INTO <table>, it would require manual execution or scheduled tasks, adding overhead and risking the timing SLA.

Memory tip: Think of Snowpipe as a "smart mailbox" - it watches your S3 bucket and automatically processes each file the moment it arrives, with no servers to babysit and no bill when nothing is loading.

No community discussion yet for this question.