DAG (EN)

Definition

A DAG (Directed Acyclic Graph) is a data structure for representing workflows, consisting of nodes (tasks) and directed edges (dependencies). The graph is acyclic, meaning there can be no cycles or backward references. In Apache Airflow, DAGs are used to model complex data pipelines and schedule their execution. Each node in the DAG represents a task that performs a specific operation.

Architecture

flowchart TD     A[Datenquelle] --> B[Extraktion]     B --> C[Transformation]     C --> D[Validierung]     D --> E[Beladung]     E --> F[Zieldatenbank]     G[Extraktion] --> H[Transformation]     H --> I[Beladung]     I --> J[Data Lake] 

Im Kontext

• Wird typischerweise zusammen mit Apache Airflow, Prefect oder Luigi verwendet
• Verwandt zu: Workflow-Management, Datenpipelines, Task-Scheduling
• Beispiel-Einsatz: ETL-Prozesse, Datenbereinigung, maschinelles Lernen