# Apache Beam

> Write a pipeline once. Run it on Flink, Spark or Dataflow.

**Field:** Data Engineering  
**Category:** Processing  
**Also known as:** Beam, Beam SDK

## Think of it like

It is like writing a song once. Any band can then play it. The words stay the same.

## Definition

Apache Beam is a free, open source way to write one pipeline that works for both batch and streaming data. You write it once in Java, Python or Go. A runner then runs it, such as Flink, Spark or Dataflow. Beam brought the ideas of event time, windows, watermarks and triggers into one model.

## Example

A team writes a Beam pipeline that cleans sensor data. It runs on Dataflow in the cloud and on Flink in their own data centre.

## Why it matters

It keeps a pipeline from being tied to one engine. Its model shaped how streaming is taught.

## Common mistake

Thinking each runner can do all of it. Some runners lack parts of Beam.

## Three principles

- **One model.** Batch and stream alike.
- **Many runners.** Flink, Spark, Dataflow.
- **Windows and triggers.** Built into the model.

## Related terms

- [Apache Flink](https://learn.benibix.com/term/data/apache-flink)
- [Apache Spark](https://learn.benibix.com/term/data/apache-spark)
- [Stream Processing](https://learn.benibix.com/term/data/stream-processing)
- [Windowing (Stream Processing)](https://learn.benibix.com/term/data/windowing)
- [Watermark (Stream Processing)](https://learn.benibix.com/term/data/watermark)

---

Source: https://learn.benibix.com/term/data/apache-beam
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
