The Metrics Query Builder provides functions to enhance analytical capabilities.
Function Types
Exclusion Functions
Exclusion functions allow you to exclude data points based on certain conditions, such as values falling outside a specified range.
- Cut Off Min — Excludes data points below a specified threshold
- Cut Off Max — Ignores values above a certain point
Clamping Functions
Clamping functions constrain data points to a specified range instead of excluding them.
- Clamp Min — Sets a floor value; any data point below the threshold is raised to the threshold
- Clamp Max — Sets a ceiling value; any data point above the threshold is lowered to the threshold
Arithmetic Functions
Arithmetic functions enable mathematical operations to customize metric calculations.
- Absolute — Returns the absolute value of data points
- Running Diff — Computes the difference between consecutive data points in a time series
- Log2 — Transforms data to a base-2 logarithmic scale
- Log10 — Transforms data to a base-10 logarithmic scale
Smoothing Functions
Smoothing functions address volatile data through moving averages and similar techniques.
- EWMA 3 — Exponentially Weighted Moving Average over 3 periods
- EWMA 5 — Exponentially Weighted Moving Average over 5 periods
- EWMA 7 — Exponentially Weighted Moving Average over 7 periods
Time Shift Functions
Time shift functions enable comparisons of data across different time periods, useful for analyzing trends, changes, or anomalies over time. Input time shift values in seconds (e.g., 3600 seconds = 1 hour).
Chain Functions
Combine two or more functions sequentially — such as applying a logarithmic scale after setting a minimum cutoff — to tailor your metric outputs precisely.
Example: Comparing Service Performance with Time Shift
The Time Shift function demonstrates practical value by comparing service call volumes across time intervals. For example, you can compare frontend service calls across a 1-hour interval to assess performance before and after deployments.
