ADA-C01 Question 358
Single answerCreate and manage resource monitors based on use cases and business requirementsA Snowflake administrator needs to control monthly credit consumption for a non-critical analytics warehouse used by several business analysts. The business requirement is: send an alert when 75% of the monthly budget is reached, and automatically stop any further warehouse usage once 100% of the monthly budget is consumed. Analysts should be able to resume the warehouse manually at the start of the next month after the monitor resets. Which configuration best meets these requirements?
- A
Create a resource monitor with a monthly frequency, assign it to the warehouse, add a NOTIFY trigger at 75%, and add a SUSPEND trigger at 100%.
- B
Create a resource monitor with a monthly frequency, assign it to the warehouse, add a NOTIFY trigger at 75%, and add a SUSPEND_IMMEDIATE trigger at 100%.
- C
Create a resource monitor with a daily frequency, assign it to the warehouse, and rely on AUTO_SUSPEND on the warehouse to stop spending when the monthly budget is reached.
- D
Create an account-level resource monitor with a monthly frequency, add a NOTIFY trigger at 75%, and add a SUSPEND trigger at 100%; this guarantees only the target warehouse is affected.
Show answer and explanation
Correct answer: A
Explanation
Resource monitors are used to track and control credit usage for warehouses based on thresholds and schedules such as daily, weekly, monthly, yearly, or never resetting. For this use case, the administrator should create a warehouse-level resource monitor with a monthly frequency and assign it to the specific analytics warehouse. Using a 75% NOTIFY trigger satisfies the alerting requirement. Using a 100% SUSPEND trigger is the best match when the goal is to prevent additional usage after the threshold is reached while allowing currently running statements to complete. In contrast, SUSPEND_IMMEDIATE would terminate running statements immediately and is better suited to stricter enforcement scenarios. Snowflake documentation and best practices distinguish warehouse-level versus account-level monitor scope and clarify the behavior of actions such as NOTIFY, SUSPEND, and SUSPEND_IMMEDIATE. AUTO_SUSPEND is unrelated to budget thresholds; it only reduces waste from idle warehouses.
- A. Correct.
Correct. A warehouse-level resource monitor can be assigned directly to the non-critical analytics warehouse, which is appropriate when the business requirement applies to that warehouse rather than the whole account. A NOTIFY trigger at 75% provides the alert threshold, and a SUSPEND trigger at 100% prevents additional warehouse usage after the current statements complete. Because the monitor resets on its monthly schedule, the warehouse can be resumed manually after the new period begins. This aligns with the requirement to stop further usage once the budget is reached without abruptly terminating running work.
- B. Incorrect.
Incorrect. SUSPEND_IMMEDIATE suspends the warehouse immediately, aborting currently running statements. That action is typically used only when the business requirement is to stop all consumption instantly, even at the cost of query failures. In this scenario, the requirement says to stop any further warehouse usage once 100% is consumed, but it does not require terminating in-flight work. SUSPEND is the better fit because it allows current statements to finish and then prevents new ones from starting.
- C. Incorrect.
Incorrect. A daily resource monitor does not match the stated monthly budget requirement. In addition, AUTO_SUSPEND only suspends a warehouse after a period of inactivity; it does not enforce a credit budget threshold or stop spending based on monthly consumption. This option confuses idle-time warehouse management with budget governance.
- D. Incorrect.
Incorrect. An account-level resource monitor affects all warehouses in the account that are not assigned to a different monitor, so it is not the right choice when the requirement is to govern one specific analytics warehouse. The statement that it guarantees only the target warehouse is affected is false. Account-level monitors are useful for broad account-wide controls, not for isolating a single warehouse unless the account design explicitly supports that approach.