dist-max

Input Stack:

query: Query

⇨

Output Stack:

TimeSeriesExpr

Compute the maximum recorded value for timer and distribution summary metrics by aggregating the max statistic. This finds the highest individual measurement value recorded within the distributions, useful for identifying peak values and worst-case scenarios.

Parameters¶

query: Query for timer or distribution summary metrics

How It Works¶

This operator automatically selects and aggregates the max statistic from timer and distribution summary metrics:

For Timers: - Aggregates the max statistic showing the longest recorded duration - Useful for identifying slowest requests or operations

For Distribution Summaries: - Aggregates the max statistic showing the largest recorded value - Useful for identifying peak sizes or amounts

Metric Requirements¶

The metrics must be instrumented using Timer or Distribution Summary classes.

Examples¶

Finding the maximum request latency:

Before	After

name,playback.startLatency,:eq	name,playback.startLatency,:eq, :dist-max

Simple usage pattern:

nf.cluster,foo,:eq,
name,http.req.latency,:eq,
:and,
:dist-max

Manual Equivalent¶

The :dist-max operator simplifies this manual query:

nf.cluster,foo,:eq,
name,http.req.latency,:eq,
:and,
statistic,max,:eq,
:and,
:max

Comparison with Other Distribution Operators¶

Operator	Statistic	Use Case
`:dist-avg`	Mean of measurements	Average performance
`:dist-max`	Maximum measurement	Worst-case scenarios
`:dist-stddev`	Standard deviation	Measurement variability

:dist-avg - Average for timer/distribution metrics
:dist-stddev - Standard deviation for timer/distribution metrics
:max - Maximum aggregation across time series (different scope)