Slide 2 of 8

The observability hierarchy

The observability hierarchy

Observability builds in layers

Sending data from external collectors gives you Level 1 (infrastructure visibility).

You can’t troubleshoot services if you don’t know your infrastructure is healthy. Getting your data into Grafana Cloud builds that foundation.

Hierarchy of Observability Needs

LevelFocusQuestion answered
Level 1Infrastructure“Is my server healthy?”
Level 2Services“Which service is the problem?”
Level 3Transactions“Why is this request slow?”
Level 4Custom logic“What’s happening in my code?”

Script

Before we dive into methods for sending data, let’s zoom out and understand where this fits in the bigger picture.

Observability isn’t one thing. It’s a hierarchy of needs.

At the foundation, Level 1, you need infrastructure visibility. Can you see if your servers, databases, and networks are healthy? This is the base everything else builds on.

At Level 2, you shift focus to services. Which service is causing the problem?

At Level 3, you dig into individual transactions. Why is this specific request slow?

And at Level 4, you add custom instrumentation for business-specific questions.

Here’s the key insight: sending data from external collectors gives you Level 1. Whether you’re using Prometheus, InfluxDB, Graphite, or OpenTelemetry, you’re building infrastructure visibility. The foundation that everything else depends on.

You can’t troubleshoot service issues if you don’t know your infrastructure is healthy. Getting your data into Grafana Cloud builds that foundation. That’s why we’re starting here.

/
Previous Next