Datastrato Enterprise
Datastrato Enterprise is a commercial distribution of Apache Gravitino. It runs the same engine, speaks the same API, and uses the same concepts, with the operational capabilities an organization needs to put a metadata lake into production.
If you have used Apache Gravitino, everything you know still applies. Metalakes, catalogs, schemas, and tables mean what they always meant. Configuration keys, REST endpoints, client libraries, and connector names are unchanged. The documentation says "Gravitino" when it means the engine and "Datastrato Enterprise" when it means the distribution and what it adds.
What Enterprise adds
Everything in Apache Gravitino 1.3.1 is included, among it tags, policies, ownership, roles and privileges, the audit log, and the Iceberg REST catalog. Datastrato Enterprise adds the following.
Identity and access. A built-in identity provider whose local users can sign in when the external identity provider is unavailable. SCIM provisioning from Microsoft Entra ID and Okta, and a server-wide directory that records where each user and group came from and can suspend and reactivate them.
More sources. Oracle, SQL Server and BigQuery catalogs, with Oracle and SQL Server support in the Trino, Spark and Flink connectors.
Governance. A governance overview that reports tag, policy and ownership coverage per catalog and per asset type, an Iceberg encryption policy checked when tables are created, search results filtered by the caller's permissions, data preview in the UI, and a bundled lineage service reached through the UI.
Operations. A supported Helm chart with license keys, request metrics recorded for every API call and shown in the UI, the MCP server deployed alongside the server, and server and audit logs kept on a persistent volume.
The administration UI, for connecting catalogs, governing assets, reviewing access, running jobs and monitoring health.
Support, with a defined response commitment and a maintained release line.
Get started
Deployment is a Helm chart pulled from a registry, installed into a Kubernetes cluster, and activated with a license key.
Start with Requirements to confirm your cluster is suitable, then Get started to install and reach a working system.
Where to go next
Connect sources covers attaching your first catalog, whether that is a lakehouse, a metastore, a relational database, or object storage.
Explore metadata covers the object model: how catalogs, schemas, tables, filesets, and models are organized within a metalake. Architecture and concepts describes how the pieces fit together.
Authorize access covers roles, users, and groups, and granting access to what you have connected.
How engines connect covers querying through Trino, Spark, Flink, and anything that speaks the Iceberg REST protocol, and the MCP server covers AI agents.
Relationship to Apache Gravitino
Apache Gravitino is a top-level project of the Apache Software Foundation. Datastrato employs many of its contributors and maintains this distribution on top of it.
The engine in Datastrato Enterprise is Apache Gravitino. Your metadata is stored in open formats, your catalogs are addressable through the standard API, and nothing in the data path is proprietary. Moving between the two, in either direction, does not require rewriting your integrations.
The upstream project documentation is at gravitino.apache.org. Where this documentation and the upstream documentation differ, this one describes what Enterprise does.