COF-C03 Question 62
Single answerStandard (Gen 1 and Gen 2)A data engineering team is loading several terabytes of historical data into Snowflake for the first time. They need broad SQL compatibility for existing ETL jobs and want to keep costs predictable without using specialized memory-optimized hardware. The architect must choose the most appropriate warehouse family for the workload. Which warehouse type should be selected?
- A
A Standard warehouse because it is designed for general-purpose workloads and provides balanced compute resources
- B
A Snowpark-optimized warehouse because it is the default choice for all large data loads and standard SQL transformations
- C
A warehouse running in economy mode because it reduces credit usage for all heavy ETL workloads without tradeoffs
- D
An external virtual warehouse because historical data loads require compute outside of Snowflake-managed resources
Show answer and explanation
Correct answer: A
Explanation
For SnowPro Core, candidates should recognize that Standard warehouses are the general-purpose warehouse family for common SQL analytics, data loading, and transformation workloads. Standard Gen 1 and Standard Gen 2 both fall into this general-purpose category, with Gen 2 representing the newer generation of standard warehouse architecture, but the key exam-level takeaway is that Standard warehouses are the right default choice unless the workload has specialized requirements. Snowpark-optimized warehouses are selected for memory-intensive workloads, not routine ETL simply because the data volume is large. This aligns with Snowflake guidance on warehouse types: choose Standard for most general compute needs and reserve specialized warehouse types for specialized workload characteristics.
- A. Correct.
Correct. Standard warehouses are intended for general-purpose workloads such as common SQL queries, ETL/ELT processing, and data loading. In both Standard Gen 1 and Standard Gen 2, they are the default fit when an organization needs broad compatibility and balanced performance rather than specialized compute characteristics.
- B. Incorrect.
Incorrect. Snowpark-optimized warehouses are intended for workloads that require more memory per node, such as certain Snowpark, machine learning, or memory-intensive processing patterns. They are not the default recommendation for ordinary large data loads and standard SQL transformations when a general-purpose warehouse is sufficient.
- C. Incorrect.
Incorrect. Snowflake warehouses do not provide an "economy mode" warehouse type for reducing cost on heavy ETL workloads. This distractor reflects a common misconception that warehouse families include tunable pricing modes beyond sizing, scaling, and warehouse type selection.
- D. Incorrect.
Incorrect. Snowflake uses Snowflake-managed virtual warehouses for compute. There is no supported concept of an "external virtual warehouse" that customers use instead of Snowflake-managed compute for routine historical data loading.