Skip to main content

Abnormal Alert

In SavingsNow, exception alerts are a vital tool for effective monitoring and management of cloud costs.

This feature allows FinOps teams to define a set of rules for automatically detecting abnormal patterns in cloud costs and sending real-time alerts. These rules are based on preset thresholds, historical data analysis, and machine learning models, accurately identifying behaviors that deviate from normal spending levels. Once abnormal costs are detected, the system automatically triggers alerts based on predefined rules, notifying relevant personnel via email (with future support for SMS or in-app notifications).

With the "Exception Alerts" feature, FinOps teams can efficiently monitor cloud costs, promptly identify potential cost risks, and take appropriate measures for adjustment and optimization. This helps businesses maintain expected spending levels, avoid unnecessary cost waste, and improve the efficiency of cloud resource utilization. It supports timely notifications through third-party integrations.

Exception Alert Rules​

The exception alert rules provide a comprehensive set of functions, including viewing, creating, editing, deleting, and detailed information.

The list supports a search function and offers both list and view display options to meet different user needs. Displayed content includes rule name, exception alert rules, status, last update time, and key information while providing operations like view, edit, and delete for easy rule management.

On the rule creation and editing page, users can set detailed parameters like rule name, description, triggering conditions, and notification methods to cater to various scenarios.

The status of exception alert rules is categorized as enabled or disabled, and users can switch the status as needed. The "last update time" shows when the rule was last modified for user awareness of changes.

Add Rule​

Click the 【New Rule】 button at the top right of the "Exception Rule Alerts" tab to open the rule creation page.

  • Specify the 【Cloud Vendor】: As basic information, the first step is to specify the cloud vendor for detection.

  • Configure the 【Detection Scope】: Multiple detection scopes can be set; click "Add Condition" to include options and select "AND" or "OR" relationships. When the number exceeds one, use the "Delete" button to reduce the scope.

    • By Cloud Account: Select from a dropdown to choose cloud accounts with detection options for "All," "Any one" (trigger alert if any cloud account hits the rule), or "Custom" (specific cloud accounts to detect).

    • By Cloud Service: Similar dropdown selection for cloud services with "All," "Any one," or "Custom" detection options.

    • By Cost Tag: Dropdown selection for tags, with the detection scope supporting only custom tag values.

  • Select 【Time Granularity】: Notification and detection timeliness options include "Daily," "Weekly," "Monthly," and "Custom Time Range."

  • Indicator Configuration for 【Consumption】: Click 【Add Condition】 to set one or more basic indicators, choosing "AND" or "OR" relationships. When the number exceeds one, use the 【Delete】 button to reduce indicators.

    • Select Indicator: Options include "Actual Spending Amount," "Actual Spending Month-on-Month Change," "Actual Spending Increase," and "Actual Spending Decrease."
    • Select Operation Type: Includes "Greater Than" and "Less Than."
    • Input Indicator Value: Specify the exact value, supporting currency type changes.

  • Indicator Configuration for 【RI/SP Metrics】: Click 【Add Condition】 to set one or more basic metrics, with similar options as above.

    • Select Indicator: Options include "RI Utilization," "RI Coverage," "SP Utilization," and "SP Coverage."
    • Select Operation Type: Includes "Greater Than" and "Less Than."
    • Input Percentage: Specify the exact percentage value.

  • Configure 【Alert Recipients】: Specify who will receive alert notifications once an exception is detected.

    • Select alert recipients from a dropdown (multi-select, optional; if none selected, no notification will be sent).
    • Alerts will be sent via email to the selected recipients.

Click the 【Confirm】 button to complete the creation of the current exception alert rule; returning to refresh the list page will show the new rule.

View Rule​

In the exception alert rule list, select a rule and click the 【View】 button to open the rule details page.

Edit Rule​

  • In the exception alert rule list, select a rule and click the 【Edit】 button.

  • In the pop-up rule editing window, make the necessary changes and click 【Confirm】 to save.

Delete Rule​

  • In the exception alert rule list, select a rule and click the 【Delete】 button.
  • Click the 【Confirm】 button to delete the current rule; the rule list refreshes to show whether the rule was successfully deleted.

Exception Alert Records​

Exception alert records display notifications generated after exception alert rules are triggered.

Based on these records, abnormal data points, events, or observations in the dataset can be identified, which deviate from normal behavior. With detection tools and programs, FinOps teams can respond quickly to maintain expected spending levels.

  • Alert List: The page provides both list and view display options, allowing users to switch as needed. Regardless of the view, detailed information such as alert source, alert time, detection scope, severity (categorized as severe or normal), abnormal costs, and abnormal trends is presented.

  • Abnormal Trends: In the list, click a record to open the details page, where trend graphs visualize specific changes in abnormalities for users to identify issues. Click 【View Rule Details】 to jump to the rule details page.