// the find
PeerDB-io/peerdb
Fast, Simple and a cost effective tool to replicate data from Postgres to Data Warehouses, Queues and Storage
PeerDB is a Postgres-to-everything CDC and batch replication tool — Snowflake, BigQuery, ClickHouse, Kafka, S3, and more. It's built specifically for teams running Postgres as the operational database who need to move data to analytics or streaming infrastructure without writing custom pipelines. Now acquired into ClickHouse Cloud as ClickPipes.
Genuinely Postgres-native: it handles TOAST columns, XMIN-based replication, jsonb/arrays, and schema changes — things that break generic ETL tools. Temporal-backed workflow orchestration means failed syncs are retried with real durability, not just a cron job hoping for the best. The SQL interface for configuring mirrors is clever — you manage replication the same way you manage tables, which means your existing migration tooling (Flyway, pgAdmin) works. E2E test coverage across Postgres, MySQL, MariaDB, MongoDB, and ClickHouse in a Tilt-based local environment is unusually thorough for an ETL tool.
AGPL license is a hard blocker for many companies — any SaaS product wrapping PeerDB has to either open-source their whole stack or negotiate a commercial license. The MinIO staging requirement for ClickHouse means you're running extra infrastructure even for a simple Postgres-to-ClickHouse setup, and the docs bury the networking gotcha (ClickHouse needs to reach MinIO directly). The project was acquired by ClickHouse and is now ClickPipes — unclear how much of the self-hosted path gets investment vs. becoming a second-class citizen. Source connectors beyond Postgres (MySQL, MongoDB) are newer and notably less battle-tested based on the connector code surface area.