COF-C03 Practice Question: Snowflake AI Data Cloud Features and Architecture
A data engineer loads a batch of records into a table and then runs a query that filters on a column with a high-cardinality value. The query profile shows partition pruning eliminated most micro-partitions, but the query still scans many rows within the surviving partitions. The table has never been clustered. Which characteristic of Snowflake micro-partitions best explains why pruning was effective at the partition level yet still left many rows to scan?
⚠ Common exam trap
The trap here is assuming that because a table is unclustered or because micro-partitions are immutable, pruning cannot occur, when in fact automatic metadata enables pruning and immutability only affects how data is rewritten.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Micro-partitions are immutable and store columnar data, so pruning uses min/max metadata while row-level filtering still requires reading surviving partitions.
Micro-partitions are immutable columnar storage units that Snowflake automatically maintains with per-column metadata, including min/max values. That metadata lets the optimizer prune partitions that cannot satisfy a filter even on unclustered tables. However, pruning is coarse: once a partition survives, its columnar data must be scanned for matching rows, which is why residual row scanning remains and why clustering can further improve pruning.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Micro-partitions are compressed and encrypted, so the optimizer cannot read metadata until the files are decompressed on the warehouse.
Why it's wrong here
Compression and encryption do not block metadata-based pruning. Snowflake maintains micro-partition metadata such as min/max values and distinct counts in Cloud Services, and pruning uses that metadata during planning, before any decompression on the warehouse occurs.
- ✗
Micro-partitions are immutable and therefore require a full table scan whenever a filter column is not the clustering key.
Why it's wrong here
Immutability does not force a full table scan. Snowflake can prune unclustered tables using the automatically collected min/max metadata for each micro-partition, which is why the query profile showed pruning even though the table was never explicitly clustered.
- ✓
Micro-partitions are immutable and store columnar data, so pruning uses min/max metadata while row-level filtering still requires reading surviving partitions.
Why this is correct
Micro-partitions are immutable columnar storage units that overlap in value ranges. Pruning uses per-column min/max metadata to skip partitions that cannot contain matching values, but within a surviving partition the columnar data must still be scanned for matching rows. This explains effective partition-level pruning with residual row scanning.
- ✗
Micro-partitions are hash-distributed across compute nodes, so pruning depends on which node holds the relevant hash bucket.
Why it's wrong here
Micro-partitions are not hash-distributed as a storage layout for pruning; pruning relies on metadata ranges stored in the Cloud Services layer. Distribution across compute nodes affects parallelism during execution, not the min/max-based elimination of micro-partitions during planning.
About these practice questions
One of 280 original COF-C03 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Snowflake exam blueprint
This COF-C03 practice question is part of Courseiva's free Snowflake certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the COF-C03 exam.