Menu
Grafana Cloud

Aggregation methods for metric queries

The following tables include the aggregation methods for the k6 metrics query type. An aggregation method is a function that returns a numerical value when applied to a specific time span of a metric.

Counter methods

MethodPrevious methodDescription
IncreaseSumCounter increase in time period
RateRPSAverage rate of counter increase per second
Cum. SumCum. Sum, CountCounter total value up to the end of time period
Cum. RateCum. Rate, Cum. RPSAverage rate from the start of test run to the end of time period

Gauge methods

MethodPrevious methodDescription
Sum (Last)Sum (Last)
Max (Last)Max (Last)

Rate methods

MethodPrevious method.Description
Non-zero fractionRateFraction of reported non-zero values. For example: successful checks
IncreaseCountCount increase over time period
Cum. SumCum. CountTotal count value up to the end of time period
RateRPSTotal count increase per second
Increase (non-zero)Passes, Non-Zero CountNon-zero observations count increase over time period
Cum. Sum (non-zero)Cum. Non-Zero CountNon-zero observations count value up to the end of time period
Rate (non-zero)Non-Zero RPSTotal count increase per second
Increase (zero)FailuresZero-observations counter increase in the time period

Trend methods

MethodPrevious methodDescription
MaxMaxMaximum observed value
MinMinMinimum observed value
AvgAverageAverage of values
Std. Dev.Std. Dev.Standard deviation
Count increaseCountCount of observations over time period
Cum. CountCum. CountTotal count of observations stored in Trend metric
Cum. MinCum. MinCumulative minimum observed value
Cum. MaxCum. MaxCumulative maximum observed value
Cum. AvgCum. Avg., Cum. MeanCumulative average of values
Cum. Std. Dev.Cum. Std. Dev.Cumulative standard deviation
MedianMedian
75th percentile75th Percentile
90th percentile90th Percentile
95th percentile95th Percentile
99th percentile99th Percentile
RateHistogram count rateAverage rate of counter increase per second