Zero-ETL Warehouse for Fast Teams

BemiDB relies on the same open-source technologies used by Netflix, Apple, and Airbnb—packaged for simple self-hosting. It connects directly to your data sources, syncing data into compressed columnar Iceberg tables stored in object storage. DuckDB powers the embedded query engine for fast analytics on large datasets. And because it uses the Postgres wire and query protocol, it works with any existing Postgres-compatible tool right out of the box.

It means you don't need separate processes to extract, transform, or load data for analytics. BemiDB automatically maps data structures, syncs data, and stores it in a query-ready columnar format—removing the overhead of building and maintaining complex pipelines.

Setup is automatic. Once you connect your data sources with a click, BemiDB detects the schema, and syncing starts in the background. No setup required on your part.

By default, all data is stored in our system to reduce data transfer costs. You can Bring Your Own S3 (BYOS3) for storing data in your own S3-compatible object storage to have full data control. Reach out to us if you have any questions, and we'd be happy to discuss your requirements.

Yes, BemiDB is an open-source project with an AGPL-3.0 license approved by the Open Source Initiative (OSI). You can use it for free, modify it, and contribute to the project. We believe that open-source is the best way to build software.

Storage in BemiDB is effectively unlimited. Data is stored in Iceberg format with compressed Parquet columnar format files on S3, which provides high scalability, availability, and durability. You won’t face hard storage limits, and performance is managed through efficient data retrieval with optimized and partitioned storage.

Contact us to learn more about our self-managed support.