ADA-C01 Question 428
Single answerMonitor Notebooks consumptionA Snowflake administrator needs to identify which users and notebooks are driving the highest Notebook compute consumption over the last 30 days so the team can implement chargeback controls. The company wants an approach that uses built-in account metadata and supports trending over time rather than a one-time UI check. Which action should the administrator take?
- A
Query the ACCOUNT_USAGE views related to Notebook executions and compute usage, then aggregate consumption by notebook owner, notebook, and date.
- B
Review QUERY_HISTORY only, because every Notebook action is represented there with enough detail to calculate all Notebook consumption directly.
- C
Use WAREHOUSE_METERING_HISTORY only, because Notebook compute is billed exclusively through virtual warehouses and can be isolated there without Notebook-specific metadata.
- D
Open Snowsight and manually inspect each notebook's recent runs, because Snowflake does not expose Notebook consumption in account metadata views.
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use ACCOUNT_USAGE metadata that captures Notebook-related activity and consumption so the administrator can build durable reporting by notebook, user, and time period. This is the most practical approach for monitoring, trending, and chargeback analysis. QUERY_HISTORY is often useful for troubleshooting SQL generated by workloads, but it is not sufficient as the primary source for Notebook consumption attribution. Likewise, WAREHOUSE_METERING_HISTORY provides warehouse-level billing data, not the notebook-level attribution needed in this scenario. Snowflake best practice for governance and monitoring is to rely on ACCOUNT_USAGE/organization metadata where available for historical analysis instead of manual UI review.
- A. Correct.
Correct. For ongoing monitoring and chargeback-style reporting, the administrator should use Snowflake's built-in account metadata in ACCOUNT_USAGE that exposes Notebook-related usage/execution details and combine that with usage aggregation over time. This approach supports trend analysis, user-level attribution, and repeatable reporting. It aligns with best practice to use account metadata views rather than relying on manual inspection.
- B. Incorrect.
Incorrect. QUERY_HISTORY can help investigate SQL statements initiated from workloads, but it is not the authoritative or complete source for Notebook consumption reporting. Notebook usage monitoring requires Notebook-specific metadata and execution/usage records rather than trying to infer all costs from query history alone. A candidate might choose this because many workload investigations start with QUERY_HISTORY, but it is insufficient for full Notebook consumption monitoring.
- C. Incorrect.
Incorrect. WAREHOUSE_METERING_HISTORY is useful for warehouse-level compute tracking, but it does not by itself provide enough Notebook-specific attribution such as which notebook or notebook owner drove the usage. Also, Notebook compute monitoring should not be reduced to warehouse metering alone when the requirement is to identify specific notebooks and users over time.
- D. Incorrect.
Incorrect. Snowsight can be useful for ad hoc inspection, but the scenario explicitly requires trend reporting over 30 days and a repeatable method using built-in account metadata. Snowflake does expose metadata views for monitoring workloads, so manual review of individual notebooks is not the best or most scalable solution.