You have a Snowflake virtual warehouse named 'REPORTING that is consistently experiencing performance issues during peak hours. You suspect that resource contention is the primary cause. Which of…

Question

Options

• AIncrease the size of from Small to Medium and enable auto-suspend after 5 minutes of inactivity.
• BEnable multi-cluster warehouse for 'REPORTING with a minimum of 2 clusters and a maximum of
• CEnable multi-cluster warehouse for with a minimum of 2 clusters and a maximum of 4 clusters. Set
• DEnable auto-suspend after 5 minutes of inactivity and implement query tagging to identify long-
• EMonitor warehouse load using Snowsight and manually increase the warehouse size when

• A
  3% (1)
• B
  11% (4)
• C
  66% (23)
• D
  3% (1)
• E
  17% (6)

Explanation

Option C is correct because resource contention means too many concurrent queries are competing for the same compute resources - the solution is scaling out (multi-cluster), not scaling up. A multi-cluster warehouse with min 2 / max 4 clusters lets Snowflake automatically spin up additional clusters during peak load, distributing concurrent queries across clusters and eliminating the bottleneck.

Why the distractors fail:

• A - Increasing warehouse size (Small → Medium) improves single-query performance but does nothing for concurrent user contention; auto-suspend is a cost tool, not a performance fix.
• B - Points in the right direction (multi-cluster) but is incomplete as stated; without proper scaling policy configuration, it doesn't fully solve the problem.
• D - Auto-suspend and query tagging are useful for cost management and diagnostics, but neither reduces contention during peak hours.
• E - Manual scaling is reactive and slow; by the time you detect the problem and resize, peak load has already degraded performance.

Memory tip: Think "contention = competition = more clusters." Whenever you see concurrent users or resource contention, the answer involves multi-cluster warehouses (scale-out). Reserve size increases (scale-up) for slow individual queries, not busy warehouses.

No community discussion yet for this question.