TroveGrid
Back to all tools
DataHub logo

DataHub

Open-source data catalog — discover, understand, and govern data across your entire stack.

0Open Source
Data Catalog
Official Website View on GitHub
Share:XLinkedInWhatsApp

Overview

DataHub is an open-source metadata platform for the modern data stack, providing data discovery, lineage tracking, schema history, and governance across your entire data ecosystem — from pipelines to dashboards.

Key Features

  • Unified metadata graph for datasets, pipelines, dashboards, and ML models
  • Auto-ingestion of metadata from Snowflake, dbt, Airflow, Kafka, and 50+ sources
  • Column-level lineage tracing from source to dashboard
  • Business glossary and data domain classification for governance
  • DataHub Actions framework for automating metadata workflows
  • Acryl Cloud for managed DataHub with enterprise SLA

Real-World Workflows

Discover and document data assets

  1. 1Deploy DataHub with Docker Compose and ingest metadata from Snowflake
  2. 2Run dbt ingestion to auto-populate model descriptions and lineage
  3. 3Data consumers search for tables by name, owner, or business term
  4. 4Tag sensitive columns and attach data owners for governance

Trace the impact of a schema change

  1. 1Search DataHub for the upstream table being changed
  2. 2View column-level lineage to find all downstream tables and dashboards
  3. 3Identify report owners who will be impacted
  4. 4Notify stakeholders before making the breaking schema change

Getting Started

# Start DataHub with Docker
python -m datahub docker quickstart

# Open UI at http://localhost:9002
# Default: datahub / datahub

# Ingest metadata from Snowflake
pip install acryl-datahub[snowflake]
datahub ingest -c snowflake_recipe.yml

Compare Alternatives

See how DataHub stacks up against similar tools.

VSopenmetadataVSamundsenVSatlasVSalation