A Data Engineer is implementing a near real-time ingestion pipeline to load data into Snowflake using the Snowflake Kafka connector. There will be three Kafka topics created. Which Snowflake objects…

Question

Options

• ATables
• BTasks
• CPipes
• DInternal stages
• EExternal stages
• FMaterialized views

• A
  87% (41)
• B
  4% (2)
• E
  6% (3)
• F
  2% (1)

Explanation

When the Snowflake Kafka connector initializes, it automatically provisions one table, one internal stage, and one pipe per Kafka topic - in this case, three of each for three topics. The internal stage acts as a temporary landing zone for micro-batched files from Kafka; the pipe (Snowpipe) continuously loads those files from the stage into the target table, enabling near-real-time ingestion without manual orchestration.

Distractors:

• Tasks (B): Tasks are Snowflake's scheduled SQL execution feature - the Kafka connector uses Snowpipe, not Tasks, for its load mechanism.
• External stages (E): The connector uses internal Snowflake-managed stages, not external stages pointing to S3/GCS/Azure.
• Materialized views (F): These are user-created precomputed query caches - the connector has no role in creating them.

Memory tip: Use the acronym TPS - Tables, Pipes, Stages (internal). Think of it as the three-layer pipeline: data lands in the Stage, gets piped through the Pipe, and is stored in the Table. If an object isn't part of that direct data-flow path (Tasks, External Stages, Materialized Views), the connector doesn't touch it.

No community discussion yet for this question.