Skip to content

Observability Metrics

Metadata

Property Value
Type data_asset
Domain technology-and-engineering
Status active
Version 1.0.0
Owner Platform Engineering
Steward Platform Engineering

Definition

Managed collection of OpenTelemetry generated metric observations produced by instrumented organizational services and infrastructure and used for operational monitoring, dashboarding, and observability analysis.

Description

The observability metrics asset contains time series data exported through OpenTelemetry from instrumented services and infrastructure. The asset supports operational dashboards, service health monitoring, infrastructure monitoring, HTTP and database performance analysis, and observability KPIs.

Source System

OpenTelemetry

Grain

One metric observation represented as a time series data point identified by its metric name, resource attributes, and metric dimensions.

Schema

OpenTelemetry metric data model exported through OTLP and stored in the observability metrics backend.

Classification

internal

Quality Expectations

- Metric names must conform to the canonical observability metric catalogue.
- Required resource attributes must be present on exported observations.
- Metric values must use the units defined by their canonical metric definitions.
- Metric dimensions must remain consistent with the field matrix.
- Metric observations should preserve service and environment identity.
- Export failures should be detectable through observability of the telemetry pipeline.

Refresh Frequency

near real-time

Retention

Defined by the observability metrics backend retention policy.

Lineage

- OpenTelemetry SDK
- OpenTelemetry Collector
- Prometheus
- Observability Dashboards
  • technology.http.request.rate
  • technology.http.request.error_rate
  • technology.http.request.latency
  • technology.http.request.active
  • technology.process.cpu.utilization
  • technology.process.memory.usage
  • technology.process.uptime
  • technology.database.operation.duration
  • technology.http.client.request.duration
  • technology.redis.operation.duration
  • technology.service.availability
  • technology.service.error_budget
  • technology.service.latency_sla_compliance
  • technology.service.performance_health

Source

This page is generated from the canonical definition definitions/data-assets/technology.observability_metrics.yaml.

Do not edit this generated page directly.