What is MyDuck Server?

MyDuck Server is an open-source analytics server that lets MySQL and PostgreSQL users run analytical queries at DuckDB speeds without leaving their familiar ecosystem. It implements both MySQL and PostgreSQL wire protocols, allowing any compatible client to connect and execute queries that are translated and executed by DuckDB's columnar OLAP engine. The server supports zero-ETL real-time replication from MySQL (via GTID-based replication) and PostgreSQL (via logical replication), making it a drop-in analytical replica. It also runs in standalone mode as a DuckDB server accessible via PostgreSQL protocol. Key capabilities include HTAP architecture support via ProxySQL or PGPool-II, end-to-end columnar I/O with Parquet/Arrow formats, bulk data loading via LOAD DATA and COPY commands, backup/restore to S3-compatible object storage, and integration with LLM applications via Model Context Protocol (MCP). The project is developed by ApeCloud and released under Apache-2.0 license. No stable releases have been published yet; the Docker image uses the 'latest' tag.

Should you self-host MyDuck Server?

MyDuck Server scores 49/100 on selfhost.directory (how scoring works) and ranks #30 of 38 in Databases. It has 594 GitHub stars.

There was no stable release in the past 12 months, so check the repository's recent activity before you rely on it.

Setup is moderate: expect some configuration beyond a single command. An official Docker image is available. It is Apache-2.0-licensed and written mostly in Go.

Its closest alternative on selfhost.directory is SurrealDB (78/100, 33 241 stars). It scores higher overall, so compare the two before you pick. See MyDuck Server vs SurrealDB side by side.