Bird
Raised Fist0
SnowflakeHow-ToBeginner · 4 min read

How to Monitor Snowflake Credits Usage Effectively

To monitor Snowflake credits, use the SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY view which tracks credit consumption by warehouses. Query this view regularly to see credit usage over time and identify cost patterns.
📐

Syntax

The main way to monitor credits is by querying the WAREHOUSE_METERING_HISTORY view in the SNOWFLAKE.ACCOUNT_USAGE schema. This view shows credit usage per warehouse per time interval.

  • WAREHOUSE_NAME: The warehouse consuming credits.
  • START_TIME: When the usage interval started.
  • CREDITS_USED: Number of credits consumed in the interval.
sql
SELECT WAREHOUSE_NAME, START_TIME, CREDITS_USED
FROM SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY
WHERE START_TIME >= DATEADD(day, -7, CURRENT_TIMESTAMP())
ORDER BY START_TIME DESC;
💻

Example

This example query shows total credits used by each warehouse in the last 7 days, helping you track which warehouses consume the most credits.

sql
SELECT WAREHOUSE_NAME,
       SUM(CREDITS_USED) AS TOTAL_CREDITS_USED
FROM SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY
WHERE START_TIME >= DATEADD(day, -7, CURRENT_TIMESTAMP())
GROUP BY WAREHOUSE_NAME
ORDER BY TOTAL_CREDITS_USED DESC;
Output
WAREHOUSE_NAME | TOTAL_CREDITS_USED ---------------|------------------- COMPUTE_WH | 125.75 ETL_WH | 98.40 REPORTING_WH | 45.20
⚠️

Common Pitfalls

Common mistakes when monitoring credits include:

  • Querying ACCOUNT_USAGE views too frequently, which can cause delays because data refreshes every few minutes.
  • Not filtering by recent dates, leading to large result sets and slow queries.
  • Ignoring warehouse suspend/resume times, which affect credit usage.

Always filter by recent time and aggregate data to get meaningful insights.

sql
/* Wrong: No date filter, large data scan */
SELECT * FROM SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY;

/* Right: Filter recent 7 days only */
SELECT * FROM SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY
WHERE START_TIME >= DATEADD(day, -7, CURRENT_TIMESTAMP());
📊

Quick Reference

FieldDescription
WAREHOUSE_NAMEName of the warehouse using credits
START_TIMEStart time of the usage interval
END_TIMEEnd time of the usage interval
CREDITS_USEDCredits consumed during the interval
CREDITS_USED_COMPUTECredits used for compute only
CREDITS_USED_CLOUD_SERVICESCredits used for cloud services
✅

Key Takeaways

Use SNOWFLAKE.ACCOUNT_USAGE.WAREHOUSE_METERING_HISTORY to track credit usage by warehouse.
Filter queries by recent dates to get timely and efficient results.
Aggregate credit usage to understand total consumption per warehouse.
Avoid querying usage views too frequently due to data refresh delays.
Monitor warehouse suspend/resume behavior to optimize credit consumption.