Observability
Observability is the ability to infer a system's internal state from its external outputs, typically metrics, logs, and traces, so engineers can diagnose unforeseen problems.
Observability is the degree to which you can understand the internal state of a system by examining the data it emits. A system is observable when engineers can ask new questions about its behavior and answer them from existing telemetry, without shipping new code to gather more data.
How It Works
Observability is commonly built on three kinds of telemetry, often called the three pillars: metrics, which are numeric measurements over time; logs, which are timestamped records of discrete events; and traces, which follow a single request as it moves through a distributed system. Modern tooling such as OpenTelemetry standardizes how this data is generated and exported. Engineers correlate the three signals to move from a high-level symptom, such as rising latency, down to a specific cause, such as a slow database query in one service.
Why It Matters
Distributed and microservice architectures fail in complex, unforeseen ways. Traditional monitoring answers known questions with predefined dashboards, but observability is about answering unknown questions, the ones you did not anticipate when you built the system. High observability shortens incident resolution, supports service level objectives, and gives teams confidence to deploy frequently. It is a prerequisite for safe continuous deployment.
Related Terms
Observability is realized through telemetry, including metrics, structured logging, and distributed tracing. It underpins service level objectives and incident management.