v.25.3New Feature

Support the histogram metric type

Official CH changelog text
Support for the histogram metric type. The interface closely mirrors the Prometheus client, where you simply call observe(value) to increment the counter in the bucket corresponding to the value. The histogram metrics are exposed via system.histogram_metrics. #75736 (Miсhael Stetsyuk).
Support for the histogram metric type in ClickHouse metrics system. The interface is designed to closely mirror the Prometheus client experience, allowing users to observe values which increment counters in corresponding buckets.

Why it matters

This feature enables users to collect and expose detailed histogram metrics, providing a granular view of value distributions over time. It solves the problem of insufficient metric types by introducing histogram support, valuable for performance monitoring and analysis. The metric data is accessible via the system.histogram_metrics table.

How to use it

Users can create histogram metrics that track data distributions by calling the observe(value) method to increment the count in the appropriate bucket. These histogram metrics are then exposed and can be queried using the system table system.histogram_metrics.

Related resources

PreviousNext