Which of the following are best practice recommendations that should be considered when loading data into Snowflake? (Select TWO).

Question

Options

• ALoad files that are approximately 25 MB or smaller.
• BRemove all dates and timestamps.
• CLoad files that are approximately 100-250 MB (or larger): This size is optimal for parallel
• DAvoid using embedded characters such as commas for numeric data types: Embedded
• ERemove semi-structured data types

• A
  8% (2)
• C
  88% (22)
• E
  4% (1)

Explanation

Snowflake recommends loading files in the range of 100–250 MB (C) because this size balances parallelism and overhead effectively. Snowflake's micro-partition architecture and parallel loading benefit most from files in this range-files that are too small create excessive metadata overhead, while files that are too large limit parallelism. Avoiding embedded delimiter characters such as commas within numeric or other data fields (D) is a best practice because such characters can break CSV/delimited file parsing and cause load errors or data corruption. Loading files of 25 MB or smaller (A) is actually discouraged because it creates too many small files and degrades performance. Removing dates/timestamps (B) and semi-structured data (E) are not recommended practices-Snowflake natively supports these data types.

No community discussion yet for this question.