A Data Engineer enables a result cache at the session level with the following command: ALTER SESSION SET USE_CACHED_RESULT = TRUE; The Engineer then runs the following SELECT query twice without dela

Question

Options

• AThe first and second run returned the same results, because SAMPLE is deterministic.
• BThe first and second run returned the same results, because the specific SEED value was
• CThe first and second run returned different results, because the query is evaluated each time it is
• DThe first and second run returned different results, because the query uses * instead of an explicit

• A
  13% (4)
• B
  78% (25)
• C
  3% (1)
• D
  6% (2)

Explanation

Option B is correct because specifying SEED(99) makes the SAMPLE function deterministic - the same seed always produces the same set of rows from the same data. Combined with USE_CACHED_RESULT = TRUE and an unchanged underlying table, Snowflake recognizes the second execution as identical to the first and returns the cached result rather than re-running the query.

Why the distractors are wrong:

• A is wrong because SAMPLE alone is not deterministic - without a SEED, it returns a different random subset on each execution. The credit here belongs to the SEED, not SAMPLE itself.
• C is wrong because the result cache prevents re-evaluation. With caching enabled, an unchanged table, and an identical query string, Snowflake serves the cached result on the second run.
• D is wrong because using * versus explicit column names has no bearing on result caching or sampling behavior - this is a red herring.

Memory tip: Think of SEED like a random number generator preset - same seed, same output every time. And Snowflake's result cache is like a photocopier: if nothing changed and the query is identical, it hands you the same page without doing the work again. SEED + result cache = identical results both runs.

No community discussion yet for this question.