Skip to content

Metrics types

Prometheus offer 4 core metric types

Counter

A counter is a prometheus metric that represents a numeric value that can increase (cumulative) o be reset to zero on restart.

Gauge

A gauge is a a prometheus metric that represents a numeric value that can increase o decrease.

Histogram

An histogram is a prometheus metric that permits to see the evolution of a metric.

Buckets

Histograms count observations in predefined buckets. Each bucket represents a range of values. Histogram buckets are useful for understanding the performance and latency of web services by providing a detailed breakdown of request durations.

<basename>_bucket{le="<upper inclusive bound>"}

This example tells asks for HTTP requests that had a duration of 0.1 seconds or less.

http_request_duration_seconds_bucket{le="0.1"}

Sum of Observations

Histograms also keep a sum of all observed values, which can be used to calculate the average

<basename>_sum

This example is tracking the total sum of the durations of all HTTP requests in seconds

http_request_duration_seconds_sum

Cumulative Counts

Histograms maintain a cumulative count of observations for each bucket. The count of events that have been observed

<basename>_count

or

<basename>_bucket{le="+Inf"}

The le stands for "less than or equal to," and "+Inf" (positive infinity) means that this bucket includes all HTTP requests, regardless of their duration.

Lets see this example

http_request_duration_seconds_bucket{le="+Inf"}
http_request_duration_seconds_count

The metric name http_request_duration_seconds_count indicates that it is tracking the total number of HTTP requests that have been observed.

It is useful for calculating the average request duration when combined with the corresponding sum metric (http_request_duration_seconds_sum). For example, dividing the sum of durations by the count gives the average duration of an HTTP request.

This gives the average duration of etcd commits called by backend

etcd_disk_backend_commit_duration_seconds_sum/etcd_disk_backend_commit_duration_seconds_count

Use the histogram_quantile() function to calculate quantiles from histograms or even aggregations of histograms. A histogram is also suitable to calculate an Apdex score. When operating on buckets, remember that the histogram is cumulative. See histograms and summaries for details of histogram usage and differences to summaries.

Summary

pending

  • Metrics types

https://prometheus.io/docs/concepts/metric_types/

  • Histograms and summaries

https://prometheus.io/docs/practices/histograms/