MotherDuck
Back to DuckDB Data Engineering Glossary
Overview
MotherDuck is a cloud-based analytics platform built on top of DuckDB that enables teams to analyze and share data without managing complex infrastructure. It combines the simplicity of DuckDB with cloud storage and computation capabilities, allowing users to work with data that exceeds their local machine's capacity.
Key Features
MotherDuck provides a hybrid execution model where queries can run locally on your machine, in the cloud, or using a combination of both. This means you can prototype queries on a sample of data locally, then seamlessly scale to the full dataset in the cloud. The service handles optimization decisions about where computation should occur automatically.
Integration
You can connect to MotherDuck through the DuckDB CLI, Python, or other supported clients using the md:
protocol prefix in your connection string. For example:
duckdb "md:my_database?motherduck_token=<token>"
The platform integrates with common data tools like dbt, Hex, and various BI platforms. It also provides a web-based SQL editor and notebook interface for interactive analysis.
Data Sharing
MotherDuck enables secure data sharing between users and teams through database snapshots. Users can create read-only copies of databases using the CREATE SHARE
command and share them via URLs, making it easy to collaborate on analysis while maintaining data governance.
Common Use Cases
MotherDuck is particularly well-suited for data teams that need to analyze moderate to large datasets without the operational overhead of managing a traditional data warehouse. It's commonly used for ad-hoc analysis, prototyping data pipelines, and building data applications that need cloud scalability with the simplicity of DuckDB.