Answer-first summary for fast verification
Answer: Produce a larger number of smaller files and process the ingestion with size Small virtual warehouses.
Snowflake's official documentation recommends splitting larger files into a greater number of smaller files to maximize load performance. This approach allows better parallelization across the compute resources in a warehouse, as the number of files that can be processed simultaneously depends on the warehouse size. Option C directly aligns with this recommendation by suggesting producing many smaller files and using Small virtual warehouses, which efficiently distributes the workload. Option A (auto-ingest Snowpipes) is designed for continuous ingestion and may not be cost-effective for large initial loads. Option B (largest files possible) contradicts Snowflake's guidance as it reduces parallel processing opportunities. Option D (batched row-by-row inserts) is inefficient compared to bulk loading methods like COPY commands. The community discussion strongly supports C with 100% consensus and references to official Snowflake documentation.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
A company is performing an initial data load of multiple terabytes into Snowflake as part of a migration. They have control over the number and size of the source CSV extract files.
What is Snowflake's recommended approach for maximizing the performance of this data load?
A
Use auto-ingest Snowpipes to load large files in a serverless model.
B
Produce the largest files possible, reducing the overall number of files to process.
C
Produce a larger number of smaller files and process the ingestion with size Small virtual warehouses.
D
Use an external tool to issue batched row-by-row inserts within BEGIN TRANSACTION and COMMIT commands.
No comments yet.