The fact that micro-partitions can overlap in their range of values helps to?

• A. prevent skew
• B. cause skew
• C. increase scan cost
• D. increase storage

Answer: (A)

Micro-partitions are small, immutable blocks of data Snowflake uses to store a table, with per-column min and max values stored for pruning. Because data isn’t strictly ordered across the whole table, the value ranges of neighboring micro-partitions can overlap. This overlap helps prevent skew by avoiding a situation where many rows with similar values all concentrate in a single partition or a tight set of partitions. With overlapping ranges, data spreads more evenly across partitions, so no single partition becomes a hotspot, which improves parallel processing and query performance. It also provides robustness as new data loads or updates shift boundaries without forcing oversized, uneven partitions. This overlapping behavior isn’t about increasing storage or inherently increasing scan cost; it mainly supports more balanced data distribution and effective pruning.