Skip to content
Search

OpenMetrics

lastmod 2025-12-03created 2025-12-026 edits

Metric Types

Counter

The Counter type is the most commonly collected metric type. It represents the cumulative count or size of the monitored events. It is typically used together with the rate function, and is often used to track the following:

  • Requests per second
  • Status codes per second
  • Request failure rate per second
  • Request success rate per second

Gauge

The Gauge type is a metric type that represents the "current state". Unlike the Counter type, it can both increase and decrease. Examples of this type:

  • Current number of connections
  • Current number of running threads
  • Current memory usage
  • Uptime

Summary

The Summary type represents metrics needed to understand system performance. Examples of this type:

  • Latency
  • Response time

Histogram

Histogram is similar to Summary, but adds the concept of "quantiles". This is very useful when reasoning about each user's experience.

Summary vs Histogram

So which should you choose, Summary or Histogram? The official documentation recommends the following.

  • If you need aggregation, use the Histogram type.
  • Also, if you have an idea of the range and distribution of the values to be observed, use Histogram.
  • Otherwise, use the Summary type.

See also

Favorite site

author Zer0 sitemap