The SQL Interface for Atlas and Enterprise Advanced (EA) lets non-developers query MongoDB data from the SQL tools they already use. It provides live, read-only access over standard ODBC and JDBC connections, eliminating the need to learn MQL or move data out of MongoDB for analytics. This simplifies operational reporting, accelerates insights, and ensures data remains accurate and consistent.
You can deploy the SQL Interface three ways: directly on an Atlas dedicated cluster, through an Atlas federated database instance, or on a self-managed Enterprise Advanced (EA) deployment. For direct SQL access to a single Atlas cluster, use a dedicated cluster. This is the recommended path for most Atlas connections, and it requires no federated database instance. Use a federated database instance when a single SQL query must span more than one data source, such as multiple clusters or S3 buckets.
Compare Deployment Options
Dedicated Cluster (Recommended for Atlas) | Federated Database Instance | On-Premises (EA) | |
|---|---|---|---|
Best for | SQL access to the data in one Atlas cluster. | A single SQL query that spans more than one data source. | SQL access to a self-managed deployment you run yourself. |
Data sources | One dedicated cluster. | Multiple Atlas clusters, Atlas Data Federation collections, and storage objects including AWS S3 buckets, Azure Blob Storage containers, Google Cloud storage buckets, HTTP URLs, and Atlas Online Archives. | One self-managed Enterprise Advanced (EA) deployment. |
Deployment requirement | Dedicated cluster, | A federated database instance. No cluster tier requirement. | A MongoDB Enterprise cluster. Community clusters are not supported. |
MongoDB version | 6.0 and later. | 5.0 and later. | 6.0 and later. |
Database edition | Atlas Core Database Edition and Atlas Infinite Database Edition. | Atlas Core Database Edition and Atlas Infinite Database Edition. | Not applicable. |
Schema management | MongoDB SQL Schema Builder CLI. | Atlas Data Federation SQL schema management, built into the Atlas UI. | MongoDB SQL Schema Builder CLI. |
Architecture | Connects directly to the cluster. | Shared and multi-tenant. The Atlas Data Federation query engine retrieves data from all underlying data sources to aggregate them. | Runs in your own environment. |
Connection and query limits | Follow the cluster itself. | The federated database instance enforces its own simultaneous connection and query limits. | Follow the deployment itself. |
All three options provide live, read-only access and use the MongoSQL dialect, which is SQL-92 compatible. You cannot use the SQL Interface to insert, update, or delete documents on any deployment, and the SQL Interface does not support other SQL dialects.
SQL Interface Components
The SQL Interface deployment consists of four components:
A JDBC or ODBC driver that connects your BI tool to your MongoDB database. The driver you select depends on your BI tool.
JSON schema are required to translate SQL queries between your BI tool and MongoDB database. For federated database instances, schema management is built into the Atlas UI. For dedicated clusters and for self-managed Enterprise Advanced (EA) deployments, you download the MongoDB SQL Schema Builder as a CLI tool. To learn more, see Schema Management.
A MongoDB database where your data resides, either an Atlas cluster or a self-managed EA database.
Key Benefits of the MongoDB SQL Interface
- Schema Improvements
- Experience significantly improved schema integration. Schema computations adapt to your data and represent your collections natively. Schemas aren't cached and you can refresh them at any time without impacting connectivity.
- Simple & Secure
- Enjoy a simplified setup with no complex proxy configurations, allowing you to connect directly to your MongoDB cluster. It offers robust authentication with support for X509 and OIDC protocols.
- First-class SQL Experience
- Leverage rich objects and arrays with native driver support, enhancing your data interactions. Server restarts are seamless, ensuring uninterrupted operations and high availability. The interface uses MongoSQL, a MongoDB-Native SQL-92 compatible dialect, which supports complex transformations like flattening nested objects and unwinding arrays directly within SQL queries.
- Unified Access
- Provides the same drivers, connectors, and schema mapping for federated database instances, dedicated clusters, and self-managed EA databases.
- BI Tool Integration
- The Power BI Connector is certified by Microsoft and the Tableau Connector is developed in partnership with Tableau. It supports Power BI with DirectQuery and Tableau Desktop, Tableau Server, Tableau Prep, and Tableau Cloud. Tableau Cloud supports SCRAM authentication only.
MongoDB SQL Interface Compatibility Considerations
When using the SQL Interface, consider the following points to ensure your setup is compatible and up-to-date:
There is no official EOL for SQL Interface components, including drivers, connectors, and the MongoDB SQL Schema Builder. Instead, their lifecycle is tied to the compatibility of the MongoDB server versions and language runtimes they are designed to work with.
SQL Interface components remain compatible with a MongoDB server version for up to 3 years after that server version's EOL. To learn more, see MongoDB Software Lifecycle Schedules.
We recommend upgrading your SQL Interface components to the latest version, especially when upgrading your MongoDB server. This ensures access to new features, performance enhancements, and long-term compatibility. To see the latest versions, go to the MongoDB download center.
Later versions of SQL Interface components might drop support for earlier MongoDB server versions. We recommend reviewing the SQL Interface Changelog to stay informed about compatibility changes and upgrade requirements.
Legacy Options
The MongoDB Connector for BI is the legacy tool for connecting BI analytics tools to a MongoDB database or cluster. The BI Connector reaches End-of-Life (EOL) in September 2026.