Size and Scale Virtual Warehouses COF-C03 Practice Questions (Page 3)

Questions 11–15

1. 11expert · hard

   A company has a multi-cluster warehouse with a min cluster count of 2 and a max cluster count of 8, using the standard scaling policy. During a flash sale, the warehouse scales to 8 clusters, but users still report slow query performance. The query profile shows that queries are not queuing, but each query is taking longer than usual. What is the most likely cause?

   • AThe queries are compute-intensive and each cluster is overloaded; scaling up the warehouse size would help.
   • BThe max cluster count is too low; increase it to 16.
   • CThe min cluster count is too high; reduce it to 1.
   • DThe scaling policy is too aggressive; change to economy.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What happens to the compute resources of a virtual warehouse when it is scaled up?

   • AEach server in the warehouse gets more compute power and memory.
   • BThe warehouse automatically adds and removes servers based on load.
   • CThe warehouse's storage capacity is increased.
   • DThe number of servers in the warehouse increases.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   Which metric from the Warehouse Load History would indicate that a warehouse is undersized for the current workload?

   • ALow average running queries.
   • BLow credit usage per query.
   • CHigh percentage of cache hits.
   • DHigh average queued time for queries.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A data analyst reports that a complex query takes 10 minutes on an X-Small warehouse. The query involves large joins and aggregations. The analyst asks if resizing the warehouse to a Small would help. What is the most accurate response?

   • AYes, but only if the query is parallelized across multiple statements.
   • BNo, resizing only affects concurrent queries, not single-query performance.
   • CYes, a Small warehouse has double the compute of X-Small, so the query will likely run faster.
   • DNo, the warehouse size only affects storage, not compute.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A team runs a warehouse that is sized as a 3X-Large. They notice that a particular query takes 5 minutes. The query profile shows that the query is not CPU-bound and does not spill to disk. The dominant cost is in the 'remote disk I/O' step. What is the most effective way to reduce the query's runtime?

   • AScale up to a 4X-Large warehouse.
   • BEnable multi-cluster scaling.
   • CChange the scaling policy to economy.
   • DOptimize the query to scan less data, such as by adding filters or using clustering keys.
   Reveal Solution Discussion Select an answer first