Unify traces, metrics, and logs with OpenTelemetry. Instrumentation, sampling, and backend-agnostic pipelines.

On this page

Observability with OpenTelemetry: Traces, Metrics, and Logs

OpenTelemetry gives you a single, vendor-neutral way to emit traces, metrics, and logs. Here’s how to adopt it.

Why OpenTelemetry #

Vendor-neutral: Switch backends (Jaeger, Prometheus, Datadog, etc.) without changing app code.
Single SDK: One instrumentation story for many languages.
Correlation: Link trace IDs to logs and metrics for faster debugging.

Quick Instrumentation (Node.js)#

javascript.javascript

const { NodeSDK } = require('@opentelemetry/sdk-node');
const { getNodeAutoInstrumentations } = require('@opentelemetry/auto-instrumentations-node');

const sdk = new NodeSDK({
  traceExporter: myExporter,
  metricReader: myMetricReader,
  instrumentations: [getNodeAutoInstrumentations()],
});
sdk.start();

Best Practices #

Sample in production (e.g. 10% of traces) to control cost and volume.
Set resource attributes (service.name, deployment.environment) for filtering.
Use semantic conventions so dashboards and alerts stay consistent.
Secure exporters (TLS, auth) and avoid PII in spans.

Start with one service and one backend; then roll out and add more signals.