Metrics overview

Chronosphere provides built-in features to help observability teams understand which metrics are driving the overall volume of data stored, and how those metrics are used.

A time series is a set of data points with matching labels. A metric is a time series with a common name, such as the __name__ label.

Chronosphere includes controls to help teams filter and aggregate metrics based on their usage. These analysis, aggregation, and filtering features are parts of the Control Plane (opens in a new tab). Understanding how Chronosphere views metrics, aggregates them, and displays them ensures you have the information you need to both find and solve problems.

Chronosphere natively supports Prometheus and StatsD metrics with the Chronosphere Collector, and OpenTelemetry metrics using the OpenTelemetry Collector.

For more information about the metric types for each client, see:

• Prometheus metrics type documentation page (opens in a new tab)
• StatsD GitHub page about metric types (opens in a new tab)
• OpenTelemetry metrics (opens in a new tab)

Analyze your usage

After you have started collecting metrics, the Metrics Live Telemetry Analyzer provides a real-time view of incoming metrics grouped by label and their relative frequency. You can use this to learn more about how often your applications emit metrics, troubleshoot spikes in ingest rates, and ensure the Collector is aware of particular metrics.

When you need to understand the value a metric is providing, the Usage Analyzer helps you find metrics that your teams find the most useful, or metrics you might consider consolidating.

Reduce data storage with aggregation rules

Chronosphere uses several aggregation rules to improve the quality and reduce the cardinality of stored metrics data. The Aggregation rules UI provides insight into your rules' performance.