Metric cardinality is the number of distinct values possible in a metric’s dimensions or labels. Higher cardinality creates more time series for a time-series database to store and query. section 2.4.3
A label that triggers cardinality explosion
An HTTP status-code label usually has only a few dozen values and is manageable. Encoding user IDs in a uid label can create tens of thousands, millions, or more possible values, making the metric expensive to store and query. section 2.4.3
Table 2.2 Overview of the costs of metrics (view table figure)
| Prometheus, OpenTelemetry collector, and OpenMetrics as emerging standard |
||
Use the cost overview to compare cardinality with metric instrumentation, agents, retention, long-term storage, and federation. section 2.4.3
