---
title: "Sampling | OpenTelemetry documentation"
description: "How to generate RED metrics and service graphs from traces on the edge with different sampling techniques."
---

> For a curated documentation index, see [llms.txt](/llms.txt). For the complete documentation index, see [llms-full.txt](/llms-full.txt).

# Sampling

Sampling reduces the cost of processing and storing large volumes of data. For example, if a service receives 100 requests per second and you set a sample rate of 10%, the system sends 10 spans per second.

Sampled data doesn’t provide a complete picture of your system, but it helps you understand service dependencies, aggregated performance, and health metrics. You can use two main sampling strategies:

- [Head sampling](head/)
- [Tail sampling](tail/)