ADA-C01 Question 540
Single answerHistorical usage dataA Snowflake administrator is asked to produce a report showing warehouse credit consumption trends for the last 10 months, broken down by warehouse and day, so finance can compare usage before and after a cost-optimization initiative. The administrator first queries the ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY view but notices that the required date range is incomplete. Which approach should the administrator use to meet the requirement with the least operational effort?
- A
Query the ORGANIZATION_USAGE.WAREHOUSE_METERING_HISTORY view, because it retains all warehouse metering history for every account for 365 days and can be used directly from any account in the organization.
- B
Query the SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY view for the available period and supplement older daily data from the Information Schema table functions, which provide longer retention for warehouse usage.
- C
Use the SNOWFLAKE database's ACCOUNT_USAGE views for historical warehouse metering, since they provide long-term retention, and if older data is still required, restore it from Time Travel on the SNOWFLAKE database.
- D
Query the SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY view, which provides long-term historical warehouse metering data, because ACCOUNT_USAGE is intended for historical reporting rather than short-term operational metadata.
Show answer and explanation
Correct answer: D
Explanation
For historical usage data in Snowflake, administrators should typically use the SNOWFLAKE database's ACCOUNT_USAGE schema rather than Information Schema when the requirement involves longer lookback periods such as many months. ACCOUNT_USAGE views are intended for account-level historical reporting, including warehouse consumption trends, query history, storage usage, and billing-related analysis. Information Schema is generally better suited to current-state or shorter-retention metadata queries. A key practical distinction is that ACCOUNT_USAGE data can have latency, but it offers the retention needed for trend analysis. ORGANIZATION_USAGE is useful for cross-account reporting from the organization account, but it is not the default choice for a single-account reporting requirement. This aligns with Snowflake documentation on ACCOUNT_USAGE and ORGANIZATION_USAGE views and best practices for usage and cost reporting.
- A. Incorrect.
Incorrect. ORGANIZATION_USAGE views are available only in the organization account and are intended for organization-wide reporting across accounts. They are not something that can be queried directly from any account in the organization. Also, using ORGANIZATION_USAGE is unnecessary if the requirement is for a single account and the needed history is available in ACCOUNT_USAGE.
- B. Incorrect.
Incorrect. Information Schema table functions are generally intended for more recent, short-retention operational metadata, not longer-term historical warehouse metering. They do not extend retention beyond what is available in ACCOUNT_USAGE for this use case. This option reflects a common misconception that Information Schema can be used to backfill older historical usage data.
- C. Incorrect.
Incorrect. Although ACCOUNT_USAGE is the correct source area for historical usage reporting, the idea of restoring older metadata from Time Travel on the SNOWFLAKE database is incorrect. System-provided metadata in SNOWFLAKE database views is not something administrators recover through Time Travel in this manner.
- D. Correct.
Correct. The SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY view is designed for historical reporting on warehouse credit usage and provides substantially longer retention than Information Schema sources. For a 10-month warehouse consumption trend report in a single account, this is the appropriate low-effort source. ACCOUNT_USAGE views are the standard choice for historical usage and governance reporting, with the understanding that they may have some latency compared to operational views.