MongoDB is a document-oriented database. It stores data as JSON-like documents and is well suited for semi-structured and evolving datasets. Etlworks reads from MongoDB, writes into it, captures document-level changes via CDC, and lets you query MongoDB with standard SQL.
What can you do with MongoDB in Etlworks?
| Use case | How |
|---|---|
| Connect to MongoDB | Create a MongoDB connection. Works against standalone servers, replica sets, and sharded clusters. Reusable across flows. |
| Extract from a collection | ETL with MongoDB. Flatten to CSV for warehouses, keep nested as JSON for queues / files. |
| Extract specific documents | Use a MongoDB Document connection to read one document or a group identified by name / wildcard / query. |
| Filter and shape with SQL | Using SQL with MongoDB — Etlworks translates standard SQL into MongoDB queries. |
| Load into MongoDB | Two patterns: one source row = one document, or whole dataset = one document. Configurable insert / update / replace strategy. |
| Copy / download / delete documents | Work with documents in MongoDB — file-style operations on MongoDB documents. |
| Capture changes (CDC) | Native log-based CDC on replica sets and sharded clusters. See Change Data Capture (CDC) from transaction log and MongoDB CDC configuration. |
Go deeper
- MongoDB connector — full configuration reference.
- ETL with MongoDB — all extract / load patterns with step-by-step setup.
- Work with documents in MongoDB — copy, download, delete documents using file-style operations.
- Using SQL with MongoDB — SQL syntax, the MongoDB-query translations Etlworks emits, and how to run native MongoDB queries or Java driver code.