Data Engineering

Data Engineering

How data actually moves between systems: batch and streaming pipelines, CDC from the write-ahead log, Kafka, Airflow, open table formats, and the quality checks that keep a green pipeline from shipping wrong data.

10 guides·Written from official documentation

How they connect

Data engineering tools and what depends on, replaces, or pairs with each - including the databases and platforms they connect to outside this category.

Choose a starting point, then follow a recorded connection.

Use the map to decide what to read next: compare alternatives, look into a listed dependency or find a complementary tool. Lanes group entries by category and sort them alphabetically, so position does not signal importance or a reading order.

These are editorial recorded associations, not live topology or universal dependencies. Each arrow runs from the entry that records a relationship to the entry it names. It does not show traffic, execution order or a verified deployment.

Follow a card to its guide or profile, or open the text view below.

  • Dependency
  • Recorded use
  • Alternatives
  • Complements

This view shows 11 technologies joined by 27 directed records.

How to read this map

Every arrow points from the entry that owns a record to the entry it references. Reciprocal records stay separate, and the line pattern identifies the field, not strength or certainty. The linked guides give the context; a record on its own does not explain technical necessity.

Dependency (dependsOn)
A lists B in dependsOn; check the guide for scope and configuration.
Recorded use (usedBy)
A lists B in usedBy. This is not converted into a dependency claim about B.
Alternatives (alternatives)
A lists B as an alternative to compare, not a guaranteed drop-in replacement.
Complements (complements)
A lists B as a complement, not a requirement to deploy both.

A filled dot leads to a guide or concept explanation; a hollow dot leads to a relationship profile. The number on a card counts its direct records in this view. Selecting a card highlights only those direct records, not a transitive dependency tree.

Technologies appear when they have at least 1 outgoing record before filtering and belong to Data Engineering. Entries from other categories appear when a listed entry names them as a dependency, use, alternative or complement. Connections to entries outside this view are not drawn, so absence here does not mean that no relationship exists.

Text view: every displayed node and recorded connection

Each sentence names the record owner first, then the referenced entry. A connection appears under both endpoints for lookup, without reversing its meaning. This view and its links work without JavaScript.

Data Engineering

Apache Airflow Data Engineering

Open guide: Apache Airflow

7 direct recorded connections in this view.

Apache Kafka Data Engineering

Open guide: Apache Kafka

9 direct recorded connections in this view.

Apache Spark Data Engineering

Open guide: Apache Spark

7 direct recorded connections in this view.

Debezium Data Engineering

Open guide: Debezium

6 direct recorded connections in this view.

  • Recorded use / usedByApache Kafka records Debezium in its usedBy list. This is a recorded use association, not an inferred dependency.
  • Dependency / dependsOnDebezium lists Apache Kafka as a dependency.
  • Dependency / dependsOnDebezium lists PostgreSQL as a dependency.
  • Dependency / dependsOnDebezium lists MySQL as a dependency.
  • Recorded use / usedByMySQL records Debezium in its usedBy list. This is a recorded use association, not an inferred dependency.
  • Recorded use / usedByPostgreSQL records Debezium in its usedBy list. This is a recorded use association, not an inferred dependency.

Database

MySQL Database

Open guide: MySQL

2 direct recorded connections in this view. 4 further records touch entries outside this view.

  • Dependency / dependsOnDebezium lists MySQL as a dependency.
  • Recorded use / usedByMySQL records Debezium in its usedBy list. This is a recorded use association, not an inferred dependency.

PostgreSQL Database

Open guide: PostgreSQL

5 direct recorded connections in this view. 30 further records touch entries outside this view.

Systems

Linux Systems

Open guide: Linux

6 direct recorded connections in this view. 31 further records touch entries outside this view.

  • Dependency / dependsOnApache Airflow lists Linux as a dependency.
  • Dependency / dependsOnApache Kafka lists Linux as a dependency.
  • Dependency / dependsOnApache Spark lists Linux as a dependency.
  • Recorded use / usedByLinux records Apache Kafka in its usedBy list. This is a recorded use association, not an inferred dependency.
  • Recorded use / usedByLinux records Apache Airflow in its usedBy list. This is a recorded use association, not an inferred dependency.
  • Recorded use / usedByLinux records Apache Spark in its usedBy list. This is a recorded use association, not an inferred dependency.

Guides

Pipelines

Streaming

Orchestration

Storage & formats

Quality

Warehousing

Compute & transformation

Schemas