DATAHUB CORE · INDEPENDENT ENGINEERING
Establish a metadata catalog and operational governance foundation using DataHub Core. Bassirah connects technical metadata with ownership, glossary terms, lineage and quality context.
Independent DataHub Core engineering.
These are independent Bassirah engineering and support offerings. They do not imply upstream vendor or foundation endorsement, official vendor support rights, enterprise feature entitlement or maintenance guarantees. Commercial DataHub offerings are not part of this service.
WHY IT MATTERS
Reliable discovery and lineage improve transparency, stewardship and change management across an enterprise data estate.
When analysts cannot find trusted datasets, they rebuild them. When engineers cannot see downstream dependencies, a schema change breaks a regulatory report. A metadata platform such as DataHub Core brings datasets, dashboards, pipelines, owners, glossary terms and lineage into one searchable place.
Technology is only half of the outcome. A catalog stays useful when ingestion runs reliably, ownership is assigned, and stewardship is part of how teams work. Bassirah engineers both sides.
Data Quality & MDM for DataHub is a Bassirah product built on open-source DataHub, Great Expectations and Splink.
WHAT BASSIRAH DELIVERS
ARCHITECTURE
A DataHub Core deployment runs several services and depends on stateful infrastructure. We design each for availability, backup and capacity, typically on Kubernetes with Helm where appropriate.
| Component | Role | Bassirah design focus |
|---|---|---|
| Web frontend and metadata service | User interface, APIs and the core metadata store logic | Sizing, SSO, TLS, horizontal scaling |
| Relational database | Primary storage for metadata | Backups, replication, restore testing |
| Search and index cluster | Search and graph-style queries over metadata | Capacity, index rebuild procedures |
| Messaging layer | Streams metadata change events between services | Retention, partitioning, monitoring |
| Ingestion framework | Pulls metadata from sources on a schedule or via push | Recipes in version control, secrets management, failure alerting |
CLEAR RESPONSIBILITIES
DataHub Core provides metadata discovery, lineage and governance workflows, but actual data authorization is generally enforced in the connected query engines, databases and security systems. Every Bassirah design defines which system is the policy decision point and which is the enforcement point.
| Concern | Where it lives | Example |
|---|---|---|
| Who may see or edit metadata in the catalog | DataHub Core roles and policies | A steward may edit glossary terms in their domain |
| Who may query the data itself | Query engines, databases, storage and IAM | Trino access control, PostgreSQL roles, object storage policies |
| How classification informs access rules | Designed process linking catalog tags to platform policies | A "restricted" tag triggers a review of engine-level rules |
We will map decision and enforcement points across your platforms.
USE CASES
Enterprise data catalog and asset ownership, so people can find trusted data and know who to ask.
Impact analysis and lineage for governed reporting, showing which reports and models a change will affect.
Domain-oriented data product discovery, with ownership, documentation and quality context per domain.
HOW WE ROLL IT OUT
WORKS WITH
Metadata from Trino/Starburst, dbt, ClickHouse, operational databases and open lakehouse sources, subject to connector support and permissions.
The full governance solution, from ownership model to lineage acceptance metrics.
DataHub Core provides metadata discovery, lineage and governance workflows, but actual data authorization is generally enforced in the connected query engines, databases and security systems. Designs must define which system is the policy decision and enforcement point.
DataHub provides ingestion connectors for many databases, query engines, transformation tools and BI platforms. Coverage and depth vary by connector and release, so we validate each required source, including lineage behavior and permissions, before committing scope.
No. Our DataHub Core services are independent engineering and support. They do not include vendor support or commercial DataHub features.
It depends on the number of sources, security integration and the governance questions you want to answer. We agree scope, sources and success criteria up front, and size the PoC accordingly.
No. A catalog improves visibility, ownership and traceability, which supports compliance work. Regulatory obligations still depend on your controls, processes and platforms, and are validated separately.
Tell us about your sources, priority domains and governance goals. We will scope a workshop or proof of concept.
Bassirah’s governed DataHub catalogue with quality gates and master data management.
Architecture Patterns
What the DataHub MCP server lets AI agents read and change, how to authenticate a shared deployment, and what semantic search requires on DataHub Core.
DataHub
