Mooncake enhances Postgres with columnstore tables and DuckDB's execution engine, enabling 1000x faster analytics without the need for complex ETL processes. This solution allows developers to efficiently query and manage large datasets stored in open formats on object storage, streamlining data analysis workflows.
Funding
Funding not disclosed
Founders
Product
Problem
Traditional data analytics often requires complex ETL processes to move data into specialized systems, adding overhead and complexity to development workflows. Analyzing large datasets typically involves significant data wrangling and infrastructure management, hindering developer productivity.
Solution
Mooncake enhances Postgres by integrating columnstore tables and DuckDB's execution engine, enabling faster analytics directly within existing Postgres environments. This allows developers to efficiently query and manage large datasets stored in open formats on object storage like S3, without extensive ETL pipelines. By leveraging familiar tools like Postgres and Python, Mooncake simplifies the data intelligence stack, allowing developers to focus on product development rather than infrastructure management. Mooncake provides a catalog for datasets and files, queryable from Postgres, and integrates with a rich Python ecosystem for data analysis and transformation.
Target Audience
Mooncake targets developers and data scientists who need to perform fast analytics on large datasets within their existing Postgres and Python environments.
Features
- Columnstore tables integrated directly into Postgres
- DuckDB execution engine for accelerated analytical queries
- Support for open table formats (Iceberg, Delta) on object storage
- Transactional inserts, updates, and deletes on columnstore tables
- Python ecosystem integration with Polars, Daft, and DuckDB
- Data catalog for managing datasets and files
- Mooncake Agents for automated data analysis and transformation (in training)