Configure volume rules to detect unexpected changes in data row counts and generate alerts when volume shifts beyond your thresholds.
Volume rules monitor data row counts and
alert you to unexpected changes. You can choose between two alert
types:
- Confidence based alerts: Trigger based on how certain the system is about detected volume changes. After five successful observer runs, alerts generate based on your configured confidence level. For more information, see Configure rule based alerts.
- Threshold based alerts: Trigger when data volume changes by a percentage you specify. For example, a 3% to 8% threshold generates alerts if volume increases or decreases within that range.
To configure volume alerts:
- Go to and select + Create Observer.
- On the Create Observer page, select your data assets and click Next.
- In the Volume row, click the gear icon in the Configure column.
-
Choose your alert type and configure settings:
- For confidence based alerts: Set the confidence level range by dragging the lower and higher value sliders, or enter percentages directly in the text boxes. Then enter the minimum change value to trigger alerts. The system calculates change from the current and previous run. For text data, change is measured in characters; for other data types, it is calculated as a percentage.
- For threshold based alerts: Click the Threshold based Alerts radio button. Select how to calculate volume change (Moving Average or Last Profile Run). In the Alert Levels text boxes, enter the lower and higher percentage values for volume change alerts. The default range is 3% to 8%.
- Click Save.
Note:
-
Confidence based alerts:
- Set the yellow range from 20% to 100% to adjust warning alerts. Below 20%, confidence is too low and generates excessive alert noise.
- For example, set the yellow range to 60% to 80% to receive warning alerts when volume change confidence falls in that range. Critical alerts trigger above 80%. No alerts trigger below 60%.
-
Threshold based alerts:
- Moving Average: Detects volume changes based on a specified number of profile runs. The default is 7 profile runs. For daily schedules, set data points to a higher number. The system analyzes data after each run and alerts you to significant volume changes. For infrequent schedules, set data points to a lower number. Less frequent runs mean longer collection times for data points.
- Last Profile Run: Detects volume changes using only the most recent profile run, reducing alert noise.
- Warning alerts trigger for changes between 3% and 8%. Critical alerts trigger for changes greater than 8%. No alerts trigger for changes less than 3%.