Open-source data cataloging
Since Sep 24Updated Nov 20, 2026Based on 21 AI answers
Reviewed by Dimitry Apollonsky
Brands AI mentions
Also mentioned rarely: Databricks Unity Catalog, Apache Gravitino, Apache Hadoop, dbt, Snowflake AI, OpenLineage, Databricks, Egeria, Open Data Discovery (ODD)
- OpenMetadataChatGPTChatGPT 100%Google AI ModeGoogle AI Mode 100%
- Apache AtlasChatGPTChatGPT 100%Google AI ModeGoogle AI Mode 100%
- AmundsenChatGPTChatGPT 64%Google AI ModeGoogle AI Mode 100%
- DataHubChatGPTChatGPT 73%Google AI ModeGoogle AI Mode 30%
- DataHubChatGPTChatGPT 18%Google AI ModeGoogle AI Mode 60%
Key takeaways for Open-source data cataloging
- Vs overall market
OpenMetadata leads Open-source data cataloging (100%), ahead of overall Enterprise Data Catalog & Governance Platforms leader Atlan. View Enterprise Data Catalog & Governance Platforms
- Across Enterprise Data Catalog & Governance Platforms
Within Enterprise Data Catalog & Governance Platforms, Atlan leads Data catalogs with automated lineage, showing AI shifts picks by buyer use case. See Data catalogs with automated lineage
Questions and answers
ChatGPTNov 9, 2026
Best open-source data cataloging tools for teams (2026)
If your goal is to help teams document, discover, understand, and govern data assets, the strongest options to evaluate are OpenMetadata and DataHub.
Both provide much more than a searchable list of databases and tables. They can help teams maintain metadata, business definitions, data ownership, technical documentation, and lineage in a centralized platform.
Here are the tools I'd shortlist, depending on your team's needs.
1. Top open-source data catalog tools
1. OpenMetadata
(Best overall starting point)
An integrated data catalog covering discovery, documentation, ownership, lineage, data quality, and governance.
- Centralized documentation for tables, columns, dashboards, and pipelines.
- Business glossary and classifications.
- Data owners, descriptions, and domain organization.
- Automated metadata ingestion from many data platforms.
- Column-level lineage and data quality features.
Best for: Teams wanting one place to document and manage their data ecosystem.
Website: OpenMetadata · GitHub ·
2. DataHub
(Best for extensibility)
A metadata platform originally developed at LinkedIn, designed for large and complex data ecosystems.
- Searchable datasets, dashboards, pipelines, and other assets.
- Rich metadata modeling and APIs.
- Ownership, tags, domains, and business glossary.
- Automated lineage and impact analysis.
- Extensible ingestion framework and event-driven architecture.
Best for: Engineering-heavy teams and organizations with complex data infrastructure.
3. Apache Atlas
(Best for Hadoop governance)
A metadata management and governance framework built around classification, relationships, and lineage.
- Metadata repository and searchable assets.
- Business and technical metadata classifications.
- Data lineage and impact analysis.
- Governance-oriented tagging.
- Integration with Hadoop ecosystem technologies.
Best for: Organizations with Hadoop, Hive, and related Apache infrastructure.
Website: Apache Atlas ·
4. Egeria
(Best for metadata interoperability)
An open metadata and governance framework focused on exchanging metadata between different tools and platforms.
- Metadata exchange across systems.
- Common metadata standards and APIs.
- Governance and stewardship capabilities.
- Federated metadata architecture.
Best for: Enterprises integrating multiple data platforms and governance systems.
Website: Egeria ·
5. CKAN
(Best for dataset publishing)
A mature open-source platform for publishing, organizing, and sharing datasets.
- Dataset descriptions and metadata.
- Searchable dataset repositories.
- Dataset categorization and tagging.
- APIs for programmatic access.
- Public and private data portals.
Best for: Government, research, and organizations managing documented dataset collections.
Website: CKAN ·
6. Marquez
(Best for pipeline lineage)
An open-source metadata service focused on tracking data jobs, datasets, and lineage through OpenLineage.
- Pipeline and dataset lineage.
- Job execution metadata.
- Upstream and downstream relationships.
- Integration with data orchestration systems.
Best for: Teams that primarily need to understand where data comes from and how it flows.
Website: Marquez ·
2. Comparison at a glance
Tool Documentation Lineage Governance Ideal use OpenMetadata Excellent Column-level Strong All-around catalog DataHub Excellent Column-level Strong Large data ecosystems Apache Atlas Good Hadoop-focused Strong Hadoop governance Egeria Good Integration-focused Strong Metadata interoperability CKAN Dataset-focused Limited Basic Dataset publishing Marquez Limited Strong pipeline focus Limited Pipeline lineage These are qualitative assessments of their core capabilities, not standardized benchmark scores.
3. Other tools worth knowing
A couple of additional projects deserve consideration:
- Amundsen: Historically popular for searchable data discovery and documentation. However, its repository was reported archived in September 2026, so I would avoid choosing it for a new production deployment.
- OpenLineage: An open standard and ecosystem for capturing lineage events. It complements cataloging platforms rather than replacing a full data catalog. Official website.
4. Which should your team choose?
My recommendations:
(First choice)
OpenMetadata
Choose this if you want a centralized platform for data documentation, descriptions, ownership, glossary terms, lineage, and quality information.
(Alternative)
DataHub
Choose this if your team has a strong data engineering function, needs extensive customization, and wants a flexible metadata platform.
CKAN
Choose this if your main requirement is organizing and publishing datasets rather than documenting a complex internal data warehouse.
Important licensing consideration: DataHub's OSS repository is Apache 2.0. OpenMetadata's main repository is also Apache 2.0, but its ingestion directory has a separate Collate Community License. Check the license of each component and version before adopting it commercially.
5. What features should you evaluate?
For a team managing data documentation, I'd prioritize these capabilities:
Capability Why it matters Table and column descriptions Explains what data actually means Business glossary Standardizes business terminology Data ownership Identifies who maintains each asset Automated metadata ingestion Reduces manual documentation Data lineage Explains where data originates and flows Search Makes documented data discoverable Version history Tracks changes to metadata Access controls Protects sensitive metadata API integration Enables documentation automation Collaboration Allows analysts and engineers to contribute One important distinction: automatically extracting schemas and metadata is not the same as documenting business meaning. Your team will still need people to define business terms, ownership, intended use, and data quality expectations.
6. A practical evaluation approach
I would run a small proof of concept with OpenMetadata and DataHub before making a decision.
Use the same representative data sources in both tools:
- Connect a database or data warehouse.
- Import tables, schemas, and column metadata.
- Add descriptions and business definitions.
- Assign data owners.
- Create a business glossary.
- Generate lineage across a few data pipelines.
- Ask analysts to search for datasets without engineering assistance.
Compare how easily each platform supports documentation, maintenance, search, and collaboration.
7. A question that would help narrow the recommendation
What kind of data documentation are you primarily trying to manage?
1. What's your primary use case?
Internal data warehouse documentation
Enterprise-wide data governance
Business glossary and data dictionary
Data lineage and pipeline documentation
Centralized documentation across multiple data sources
2. What's your data infrastructure?
Snowflake / BigQuery / Databricks
PostgreSQL / MySQL / SQL Server
AWS / Azure / Google Cloud
Multiple databases and platforms
Not decided yet
3. Deployment preference
Self-hosted / fully open source
Open source with optional commercial support
Cloud-hosted is acceptable
[Get a tailored recommendation]
I can then narrow this down to the best two or three tools and compare their documentation capabilities, integrations, deployment requirements, and licensing.
Pages cited in this answer 23
- open-metadata.orgOpenMetadata: #1 Open Source Context Layer
- github.comGitHub - open-metadata/OpenMetadata: The Open Context Layer for Data and AI , OpenMetadata is the open platform for building trusted data context and business semantics for humans, AI assistants, and agents. · GitHub
- docs.open-metadata.orgOpenMetadata Documentation - OpenMetadata Documentation
- datahubproject.ioDataHub | AI & Data Context Management Platform
- github.comGitHub - datahub-project/datahub: The Context Platform for your Data and AI Stack · GitHub
- docs.datahub.comWhat is DataHub? | DataHub
- atlas.apache.orgApache Atlas – Data Governance and Metadata framework for Hadoop
- github.comGitHub - apache/atlas: Apache Atlas - Open Metadata Management and Governance capabilities across the Hadoop platform and beyond · GitHub
- egeria-project.orgEgeria Project - Open metadata and governance for enterprises
- github.comGitHub - odpi/egeria: Project Egeria's core platform providing Open Metadata and Governance support. It has an open type system, distributed knowledge graph, REST APIs, intelligent governance and pre-built content · GitHub
- ckan.orgCKAN - The open source data management system
- github.comGitHub - ckan/ckan: CKAN is an open-source DMS (data management system) for powering data hubs and data portals. CKAN makes it easy to publish, share and use data. It powers catalog.data.gov, open.canada.ca/data, data.humdata.org among many other sites. · GitHub
- marquezproject.aiMarquez Project | Marquez Project
- github.comGitHub - MarquezProject/marquez: Collect, aggregate, and visualize a data ecosystem's metadata · GitHub
- datacatalog.netBest Open-Source Data Catalogs: A Practical Shortlist | DataCatalog.net
- faun.devBest open source Data Catalogs: ranked and compared
- atlan.comOpen Source Data Catalog: Top 5 Tools To Consider in 2026
- github.comdatahub/docs/features.md at master · datahub-project/datahub · GitHub
- datatrail.aiOpen Source Data Catalog Tools in 2026 | DataTrail
- deployopen.comOpen-source data catalogs compared: OpenMetadata, DataHub, Atlas
- openlineage.ioHome | OpenLineage
- github.comGitHub - open-metadata/OpenMetadata: The Open Context Layer for Data and AI , OpenMetadata is the open platform for building trusted data context and business semantics for humans, AI assistants, and agents. · GitHub
- github.comOpenMetadata/ingestion/LICENSE at main · open-metadata/OpenMetadata · GitHub
Google AI ModeNov 9, 2026
Managing data documentation and discovery becomes much smoother with open-source tools, especially for engineering-heavy or budget-conscious teams.
The top open-source data cataloging tools available right now stand out across distinct use cases:1. DataHub (LinkedIn)
- Best for: Comprehensive data discovery, end-to-end lineage, and active governance.
- Overview: Originally built at LinkedIn and now an independent, highly active open-source project, DataHub is one of the most feature-complete modern catalogs.
- Key Strengths:
- Rich search and discovery across datasets, dashboards, ML models, and unstructured files.
- Deep automated data lineage tracking (column-level and pipeline-level).
- Built-in support for data contracts, PII tagging, and basic observability.
- Extensible graph-based architecture with rich REST/GraphQL APIs and UI-based ingestion.
2. OpenMetadata
- Best for: A unified "knowledge graph" approach combining semantics, lineage, and AI context.
- Overview: Designed as an open-standard metadata and governance platform, OpenMetadata has rapidly grown into a massive community project.
- Key Strengths:
- Combines three primitives into a single graph: Metadata (tables, pipelines, models), Meaning (glossaries, metrics, W3C standards), and Memory (collaboration, audit logs, and feedback loops).
- 130+ out-of-the-box data connectors.
- Strong focus on data quality, ownership assignment, and policy-driven governance.
- Modern, intuitive React-based user interface.
3. Apache Atlas
- Best for: Enterprise-grade Hadoop, spark, and complex hybrid-cloud governance.
- Overview: Incubated originally under the Hortonworks/Apache ecosystem, Apache Atlas is a veteran framework trusted in heavy corporate and big-data environments.
- Key Strengths:
- Deep integration with security and access frameworks like Apache Ranger (for dynamic masking and access control).
- Flexible type and entity system to define custom metadata models.
- Comprehensive classification and tag-propagation mechanisms.
- Note: The UI can feel heavier and more complex to configure than newer alternatives.
4. Amundsen (Lyft)
- Best for: Lightweight, high-speed data discovery and search.
- Overview: Created at Lyft, Amundsen focuses strictly and cleanly on the "Google-like search" experience for data assets.
- Key Strengths:
- Blazingly fast search interface built on Elasticsearch/Neo4j.
- Great at showing "popular tables" based on query logs (helping analysts see what people actually use).
- Less bloated if you only need search and basic table/column documentation rather than heavy workflow governance.
- Note: Features a smaller scope of out-of-the-box active governance compared to DataHub or OpenMetadata.
5. Unity Catalog (Open Source / Databricks Ecosystem)
- Best for: Lakehouse architectures centered around Delta Lake and open table formats.
- Overview: Databricks open-sourced Unity Catalog OSS , making it a standard choice if your data platform heavily leverages Parquet, Delta Lake, and open data/AI formats.
- Key Strengths:
- Centralized access control, table/column governance, and lineage natively tied to storage layers.
- Built from the ground up for modern data lakehouse interoperability.
Quick Comparison Matrix
Tool Primary Origin Best Feature UI / Ease of Use DataHub LinkedIn Full-stack governance & lineage Modern & extensible OpenMetadata Community / Open Unified knowledge graph + AI context Very clean & collaborative Apache Atlas Apache Foundation Deep security/Ranger integration Enterprise / Complex Amundsen Lyft Fast, simple dataset search Minimalist Unity Catalog Databricks Lakehouse & Delta/Parquet integration Developer/Engine-focused To help narrow down the best fit, tell me:
- What data warehouse or storage tools are you currently using (e.g., Snowflake, BigQuery, Postgres, S3/Delta)?
- Are you looking primarily for basic table/column documentation or heavy compliance/lineage tracking?
ChatGPTNov 3, 2026
Top open-source data cataloging tools (2026)
If your team's goal is to centralize data documentation, make datasets easier to discover, document tables and columns, track ownership, and understand data lineage, there are several excellent open-source platforms worth evaluating.
My top two recommendations are OpenMetadata and DataHub. Both are actively developed, have substantial communities, and offer the core capabilities needed for modern data documentation and governance.
Here's how I'd compare the leading options.
1. Top five options
1. OpenMetadata
(Best overall starting point)
Apache 2.0 license
- Centralized documentation for databases, tables, columns, dashboards, and pipelines.
- Business glossary, data ownership, tagging, and descriptions.
- Data lineage, data quality, and profiling.
- Broad integrations with modern data platforms.
Why consider it: Particularly strong if you want documentation, discovery, quality, and governance in one platform.
2. DataHub
(Best for extensibility)
Apache 2.0 license
- Searchable catalog for datasets, dashboards, pipelines, and other assets.
- Rich metadata management and documentation.
- Column-level lineage and ownership tracking.
- Strong APIs and automated metadata ingestion.
Why consider it: Excellent for engineering-heavy teams that want to automate documentation and integrate cataloging into their data workflows.
3. Apache Atlas
Apache 2.0 license
- Metadata management and classification.
- Data lineage and governance.
- Business and technical metadata.
- Particularly relevant to Hadoop-based environments.
Why consider it: A good candidate if your organization has established Hadoop infrastructure or more traditional enterprise governance requirements.
4. Open Data Discovery (ODD)
Open source
- Metadata discovery and collection.
- Data quality and observability capabilities.
- Business metadata and glossary support.
- Useful for data mesh and data-contract-oriented workflows.
Why consider it: Worth investigating if your documentation initiative is closely connected to data quality and data contracts.
5. Marquez
Apache 2.0 license
- Tracks datasets, jobs, and data transformations.
- Supports the OpenLineage standard.
- Helps explain where data originated and how it changes.
Why consider it: Best viewed as a complementary lineage solution rather than a complete documentation catalog.
This shortlist reflects 2026 project activity, feature coverage, and community adoption, rather than treating GitHub stars as the only measure of quality. OpenMetadata and DataHub lead most general-purpose comparisons.
2. Side-by-side comparison
Capability OpenMetadata DataHub Atlas ODD Marquez Dataset documentation Excellent Excellent Good Good Limited Column descriptions Yes Yes Yes Yes Limited Business glossary Yes Yes Yes Yes No Data lineage Yes Yes Yes Partial Excellent Data quality Built-in Supported Limited Yes No Automated metadata ingestion Yes Yes Yes Yes Yes Self-hosted Yes Yes Yes Yes Yes Feature coverage is based on published project documentation and comparative research; integration depth varies by data source.
3. What about other popular options?
A couple of other projects deserve consideration:
- Amundsen: Historically popular for data discovery, but its repository was archived in September 2026. I wouldn't recommend starting a new implementation with it.
- Apache Gravitino: Interesting for teams building a unified metadata and data-access architecture, particularly across multiple data systems. Official website.
4. My recommendation
For a team primarily interested in managing data documentation, I'd narrow the evaluation to these two:
OpenMetadata (My first choice)
Choose this if you want:
- A relatively comprehensive, integrated catalog.
- Easy-to-understand documentation and discovery.
- Business glossaries and data ownership.
- Data quality and lineage alongside documentation.
DataHub
Choose this if you want:
- A highly extensible metadata platform.
- Strong programmatic integrations.
- Automated metadata ingestion.
- Advanced lineage and engineering workflows.
One important consideration: open source doesn't mean zero operational cost. Both platforms require infrastructure, deployment, upgrades, connector configuration, and ongoing metadata stewardship.
5. A few questions to narrow down the best fit
The right choice depends heavily on your existing data infrastructure and how your team plans to maintain documentation.
1. What data platforms do you currently use?
Snowflake / BigQuery / Databricks
PostgreSQL / MySQL / SQL Server
AWS / Azure / Google Cloud
Multiple platforms and data warehouses
Other / not sure
2. What is your primary objective?
Centralized documentation and searchable data dictionary
Data lineage and impact analysis
Data governance, ownership, and business glossary
All of the above
3. How would you prefer to deploy it?
Self-hosted (our own infrastructure)
Cloud-hosted
Either is fine
4. Approximately how many people will use the catalog?
Fewer than 10
10–50
51–200
More than 200
[Find our best-fit solution]
With those answers, I can narrow this down to the best two or three platforms and suggest a practical implementation approach.
Pages cited in this answer 18
- open-metadata.orgOpenMetadata: #1 Open Source Context Layer
- github.comGitHub - open-metadata/OpenMetadata: The Open Context Layer for Data and AI , OpenMetadata is the open platform for building trusted data context and business semantics for humans, AI assistants, and agents. · GitHub
- datahubproject.ioDataHub | AI & Data Context Management Platform
- github.comGitHub - datahub-project/datahub: The Context Platform for your Data and AI Stack · GitHub
- atlas.apache.orgApache Atlas – Data Governance and Metadata framework for Hadoop
- github.comGitHub - apache/atlas: Apache Atlas - Open Metadata Management and Governance capabilities across the Hadoop platform and beyond · GitHub
- opendatadiscovery.orgOpenDataDiscovery Platform - The First Tool to Provide Truly End-to-End Data Discovery, Observability, and Trust
- github.comOpen Data Discovery · GitHub
- marquezproject.aiMarquez Project | Marquez Project
- github.comGitHub - MarquezProject/marquez: Collect, aggregate, and visualize a data ecosystem's metadata · GitHub
- datatrail.aiOpen Source Data Catalog Tools in 2026 | DataTrail
- datacatalog.netBest Open-Source Data Catalogs: A Practical Shortlist | DataCatalog.net
- atlan.comOpen Source Data Catalog: Top 5 Tools To Consider in 2026
- faun.devBest Data Catalogs: ranked and compared
- github.comGitHub - opendatadiscovery/awesome-data-catalogs: 📙 Awesome Data Catalogs and Observability Platforms. · GitHub
- deployopen.comOpen-source data catalogs compared: OpenMetadata, DataHub, Atlas
- atlan.com17 Best Data Catalog Tools in 2026: Buyer's Guide
- gravitino.apache.orgWelcome | Apache Gravitino
Google AI ModeNov 3, 2026
When looking for top-rated open-source data catalog options to manage data documentation, discovery, and lineage, the market is largely led by
a few robust, highly active projects.The leading open-source data catalog options vary by architectural fit and specific team needs:
- DataHub
- Best for: Modern data stacks and microservices-heavy architectures.
- Key Strengths: Originally built at LinkedIn, DataHub features a powerful metadata graph, rich UI, deep search capabilities, automated data lineage (including column-level lineage), and robust extensible connectors for modern cloud data warehouses, BI tools, and orchestration frameworks.
- OpenMetadata
- Best for: Centralizing governance, documentation, and collaboration out-of-the-box.
- Key Strengths: Built around a central metadata standard with a heavy focus on data collaboration, data quality, and ownership tracking. It features an intuitive user interface, extensive built-in connectors (>120+), automated profiling, and discussion threads tied directly to data assets.
- Apache Atlas
- Best for: Enterprise Hadoop or big-data ecosystem integrations.
- Key Strengths: Traditionally incubated in big data ecosystems (Hive, HBase, Spark), Atlas excels at tag/classification propagation and deep governance frameworks. However, its UI and operational setup can feel heavier and more complex compared to newer alternatives.
- Amundsen
- Best for: Page-rank style data discovery focused heavily on data analysts and consumers.
- Key Strengths: Originally open-sourced by Lyft, Amundsen focuses strictly on data discovery and metadata search. It is lightweight and clean, though it leans less on deep active governance and more on simple, searchable data documentation.
If you want to narrow this down, tell me:
- What data stack/warehouses you are currently using (e.g., Snowflake, BigQuery, Postgres, dbt)
- Whether you need deep automated data lineage or just clean text documentation/tagging
I can help recommend the best fit and outline what the maintenance/deployment effort looks like.
Google AI ModeOct 21, 2026
Top open-source data catalog platforms for organizing team documentation include
Apache Atlas, Amundsen, DataHub, and OpenMetadata, each offering robust metadata management and search capabilities.Open-Source Data Catalog Tools
- Apache Atlas: Provides deep enterprise governance, lineage tracking, and integration with the Hadoop ecosystem.
- Amundsen: Built by Lyft, it features a page-rank style search and strong discovery features focused on data usability.
- DataHub: Developed originally at LinkedIn, it offers high scalability, rich metadata types, and strong operational lineage.
- OpenMetadata: Uses a modern user interface, built-in collaboration tools, and automated metadata ingestion pipelines.
Feature Comparison
Platform Best For Key Strengths Apache Atlas Enterprise Hadoop/Hive ecosystems Deep governance and security integrations Amundsen Fast data discovery and search Clean user experience and active community DataHub Large-scale data environments Extensible architecture and strong lineage OpenMetadata Collaborative team documentation Automated ingestion and built-in workflows Would you like to know more about the deployment requirements or integration support for any of these specific platforms?
ChatGPTOct 21, 2026
If your goal is to help teams keep data documentation organized, searchable, and up to date, I’d focus on open-source data catalog platforms that combine metadata discovery, business glossaries, ownership, lineage, and collaborative documentation.
The best choice depends on whether your team primarily documents databases and analytics assets, builds a searchable internal knowledge base, or needs governance across a large data platform.
Open-source data catalog platforms worth evaluating
1. OpenMetadata
(Best all-around starting point)
An integrated catalog for dataset descriptions, ownership, business glossaries, lineage, data quality, and governance. It helps teams keep technical metadata and business definitions together rather than scattered across documents.
Best for: Teams that want one shared place to document and understand their data assets.
Website ·
2. DataHub
(Best for engineering-heavy teams)
A flexible metadata platform with searchable datasets, dashboards, pipelines, ownership, lineage, and governance. Its metadata graph is particularly useful when you want to connect documentation to the systems and assets that produce or consume data.
Best for: Data engineering teams with multiple warehouses, BI tools, and pipelines.
Website ·
3. Apache Atlas
A metadata and governance framework built for managing data classifications, lineage, and governance policies. It is worth considering if your organization already uses Hadoop-oriented infrastructure and needs structured governance.
Best for: Enterprise data governance, especially in Hadoop-centric environments.
Website ·
4. Marquez
An open-source metadata service focused on data lineage and pipeline execution metadata, using the OpenLineage standard. It helps document where datasets come from and which jobs produce or consume them.
Best for: Teams whose main documentation gap is pipeline provenance and lineage.
Website ·
5. CKAN
An open-source data portal for publishing, organizing, and discovering datasets with metadata. It is a strong fit for curated collections of datasets, especially when documentation needs to be accessible to users outside the data engineering team.
Best for: Public data portals, research organizations, and shared dataset repositories.
Website ·
Quick comparison
Platform Documentation strengths Setup profile OpenMetadata Glossaries, descriptions, ownership, lineage, quality Moderate DataHub Search, metadata relationships, governance, lineage Moderate–high Apache Atlas Classifications, policies, governance, lineage Higher for new deployments Marquez Pipeline provenance and lineage More focused scope CKAN Dataset descriptions and searchable data portals Moderate These setup profiles are general guidance, not measured installation benchmarks. All five are open-source projects, but the breadth of features and maintenance requirements differ.
My recommendation
- Start with OpenMetadata if you want a central catalog that makes documentation easier to create, maintain, and discover.
- Choose DataHub if your team needs a highly extensible metadata platform integrated with a complex data stack.
- Choose CKAN if you're cataloging datasets for a broad audience rather than primarily documenting internal analytics infrastructure.
- Add or consider Marquez if lineage is your main requirement.
One practical note: Amundsen used to be a popular alternative, but its repository was archived in September 2026, so I would not prioritize it for a new deployment.
When evaluating your shortlist, check that the platform supports your actual data sources, allows teams to edit descriptions and glossary terms, and provides a sustainable way to keep metadata refreshed. Self-hosting also means planning for upgrades, access controls, and connector maintenance.
One question to help narrow this down: What kind of documentation are you mainly trying to organize?
1. What do you need to catalog?
Tables, schemas, pipelines, and dashboards
Business definitions, glossaries, and team knowledge
Both technical and business documentation
Research or public datasets
2. How do you want to deploy it?
Self-hosted, open-source deployment
Managed hosting is acceptable
Either option
3. What's your team size?
Small team (1–10)
Medium team (11–50)
Large organization (50+)
[Find the best fit]
Pages cited in this answer 18
- github.comGitHub - open-metadata/OpenMetadata: The Open Context Layer for Data and AI , OpenMetadata is the open platform for building trusted data context and business semantics for humans, AI assistants, and agents. · GitHub
- open-metadata.orgMetadata Platforms: Use Cases, Components, and 7 Notable Solutions | OpenMetadata Learning Center
- github.comGitHub - opendatadiscovery/awesome-data-catalogs: 📙 Awesome Data Catalogs and Observability Platforms. · GitHub
- open-metadata.orgOpenMetadata: #1 Open Source Context Layer
- github.comGitHub - open-metadata/OpenMetadata: The Open Context Layer for Data and AI , OpenMetadata is the open platform for building trusted data context and business semantics for humans, AI assistants, and agents. · GitHub
- docs.datahub.comWhat is DataHub? | DataHub
- github.comdatahub/README.md at master · datahub-project/datahub · GitHub
- datahub.comDataHub | AI & Data Context Management Platform
- github.comGitHub - datahub-project/datahub: The Context Platform for your Data and AI Stack · GitHub
- datacatalog.netBest Open-Source Data Catalogs: A Practical Shortlist | DataCatalog.net
- dataworkers.ioHow to Evaluate an Open-Source Data Catalog | Data Workers
- atlas.apache.orgApache Atlas – Data Governance and Metadata framework for Hadoop
- github.comGitHub - apache/atlas: Apache Atlas - Open Metadata Management and Governance capabilities across the Hadoop platform and beyond · GitHub
- marquezproject.aiMarquez Project | Marquez Project
- github.comGitHub - MarquezProject/marquez: Collect, aggregate, and visualize a data ecosystem's metadata · GitHub
- ckan.orgCKAN - The open source data management system
- github.comGitHub - ckan/ckan: CKAN is an open-source DMS (data management system) for powering data hubs and data portals. CKAN makes it easy to publish, share and use data. It powers catalog.data.gov, open.canada.ca/data, data.humdata.org among many other sites. · GitHub
- github.comGitHub - amundsen-io/amundsen: Amundsen is a metadata driven application for improving the productivity of data analysts, data scientists and engineers when interacting with data. · GitHub
Open-source data cataloging: common questions
- Which brands does AI recommend for Open-source data cataloging?
- AI most often mentions OpenMetadata (100%), Apache Atlas (100%), Amundsen (81%), DataHub (52%) and DataHub (38%) for Open-source data cataloging in Enterprise Data Catalog & Governance Platforms.
- Does the Open-source data cataloging leader match the overall Enterprise Data Catalog & Governance Platforms market?
- OpenMetadata leads Open-source data cataloging (100%), ahead of overall Enterprise Data Catalog & Governance Platforms leader Atlan.
- What questions do buyers ask AI about Open-source data cataloging?
- Common buyer questions include "What are the best open-source data cataloging tools available for teams to manage data documentation?", "What are the top-rated open-source options for data cataloging? Our team is looking for something to help us manage our data documentation.", "Can you recommend some open-source platforms for data cataloging that help teams keep their documentation organized?".
For agents: this page as Markdown, https://parse.gl/topics/open-source-data-cataloging.md. JSON, MCP · free for the last 90 days, checks from $0.01
This page
Read and search
Connect
An agent can save work before signup: POST https://agent.parse.gl/v1/workspaces
What it costs
Prices: https://parse.gl/pricing.md
Map of the site: https://parse.gl/llms.txt