COF-C03 Question 83
Single answerWorkloadsA retail company runs hourly ELT jobs that load and transform point-of-sale data into reporting tables. At the same time, business analysts run ad hoc dashboard queries against the same data warehouse. During business hours, analysts report slow and inconsistent query performance whenever the hourly ELT job starts. The Snowflake administrator wants to reduce contention between these two workloads while keeping the solution simple to operate. Which action should the administrator take?
- A
Create a separate virtual warehouse for the ELT jobs and leave analyst queries on their existing warehouse
- B
Convert the existing standard table storage to external tables so compute contention is reduced
- C
Enable Time Travel retention to a longer period so analyst queries can avoid ELT impact
- D
Increase the maximum number of columns allowed in the reporting tables so queries scan less data
Show answer and explanation
Correct answer: A
Explanation
Snowflake best practice for mixed workloads is to isolate them using separate virtual warehouses. Warehouses do not share compute resources, so assigning ELT jobs to one warehouse and BI or ad hoc analyst queries to another avoids resource contention and improves performance consistency. This is one of Snowflake's core workload management patterns. If needed, each warehouse can also be independently resized or configured for auto-suspend and auto-resume to balance cost and performance. By contrast, features such as Time Travel and external tables serve different purposes and do not solve compute contention. Relevant Snowflake documentation covers virtual warehouses, workload isolation, and warehouse-based compute separation for concurrent workloads.
- A. Correct.
Correct. In Snowflake, virtual warehouses provide independent compute resources. Separating ELT processing and interactive analytics onto different warehouses is a standard workload isolation pattern that prevents the two workloads from competing for the same compute resources. This improves concurrency and predictability while remaining operationally simple.
- B. Incorrect.
Incorrect. External tables are used to query data stored in external cloud storage and do not address compute contention between concurrent workloads running in Snowflake virtual warehouses. This option confuses storage architecture with compute isolation.
- C. Incorrect.
Incorrect. Time Travel allows access to historical data versions for recovery and auditing use cases. Extending Time Travel retention does not isolate workloads or improve concurrent query performance during ELT execution.
- D. Incorrect.
Incorrect. The number of columns allowed in a table is unrelated to workload contention. Query performance issues in this scenario are caused by compute resource sharing between ELT and analyst queries, not by a table definition limit.