Skip to content

Metrics

A Metric represents the actual monitoring data collected from a monitored resource.

Metrics are time-series measurements associated with a Metric Type and produced by monitoring probes.

They represent the real values observed in the system over time.

Relationship with Metric Types

Metrics are always associated with a Metric Type, which defines the structure and meaning of the measurement.

The relationship between the entities is therefore:

flowchart TD

O["Object<br>(monitored resource)"]
MT["Metric Type<br>(measurement definition)"]
M["Metric<br>(time-series values)"]

O --> MT
MT --> M

While Metric Types define what can be measured, Metrics represent the values that have been measured.

For example:

Object Metric Type Metric Value
Server A CPU Usage 62%
Server A CPU Usage 58%
Server A CPU Usage 71%

For a conceptual explanation of this model, see
Metrics and Metric Types.

Time-Series Data

Metrics are stored as time-series data.

Each measurement includes:

  • a timestamp
  • a value or status
  • the metric it belongs to

Example sequence:

14:00 → 55%
14:05 → 62%
14:10 → 58%
14:15 → 61%

These time-series measurements allow the platform to analyze trends and detect anomalies.

Metrics can represent two main data types:

Value Metrics

Numerical measurements such as:

  • CPU utilization
  • network traffic
  • latency
  • bandwidth usage

These metrics are typically visualized as time-series charts.

Status Metrics

State measurements representing the condition of a service or component.

Examples include:

  • service status (OK / Warning / Critical)
  • connectivity state
  • operational state

These metrics are typically displayed as tables or status indicators.

Viewing Metric Data

Metric values can be accessed from the Tree Hierarchy View.

Each metric node includes an action button:

Metric Data

Selecting this action opens a dialog displaying the historical data for the metric.

Depending on the metric type, the interface shows:

  • a chart for numerical metrics
  • a table for status metrics

This dialog allows users to analyze the behavior of the metric over time.

Operational Actions

Metrics support several operational actions that allow administrators to control monitoring behavior.

These include:

  • Downtimes – temporarily suspend alerts related to the metric
  • Dispatchers – configure automated responses triggered by metric conditions

The interface also supports mass operations, allowing the same configuration to be applied to multiple metrics:

  • Massive Downtime
  • Massive Dispatcher

These operations are available when multiple metrics are selected.

Connections

Metrics can be linked to additional operational entities through the Connections View.

Supported connections include:

  • Services – associate the metric with a monitored service
  • Downtimes – schedule maintenance windows
  • Dispatchers – configure automated actions triggered by metric conditions

These relationships allow metrics to participate in the operational monitoring and automation workflows of the platform.

Role of Metrics in the Platform

Metrics are the fundamental data produced by the monitoring system.

They provide the raw information used to:

  • visualize infrastructure performance
  • detect anomalies
  • monitor service health
  • trigger automated actions
  • feed dashboards and widgets

All analytical and operational views in the platform ultimately rely on metric data collected from monitored objects.