Which of the following is an example of an operation that can be completed without requiring compute, assuming no queries have been executed previously?

Question

Options

• ASELECT SUM (ORDER_AMT) FROM SALES;
• BSELECT AVG(ORDER_QTY) FROM SALES;
• CSELECT MIN(ORDER_AMT) FROM SALES;
• DSELECT ORDER_AMT * ORDER_QTY FROM SALES;

• A
  3% (1)
• B
  90% (36)
• C
  3% (1)
• D
  5% (2)

Explanation

Snowflake's Cloud Services layer maintains column-level metadata statistics for each micro-partition, including row count, null count, min, max, and aggregate values for numeric columns (such as sum and count). For AVG(ORDER_QTY), Snowflake can derive the result from precomputed metadata (sum divided by count) without engaging a virtual warehouse. SUM(ORDER_AMT) and row-level expressions like ORDER_AMT * ORDER_QTY require scanning actual data. While MIN is also a metadata-available statistic, AVG(ORDER_QTY) is the answer the exam identifies as leveraging Snowflake's metadata optimization to avoid compute costs when the result cache is empty.

No community discussion yet for this question.