Postgres + ClickHouse: Building a Real-Time Open-Source Data Stack

​Description
In this instructor-led, hands-on workshop, you'll learn how to integrate Postgres and ClickHouse to build a high-performance, open-source data stack that supports real-time applications.

​Abstract
An increasing number of organizations are adopting a proven architectural pattern: using PostgreSQL for transactional workloads and ClickHouse for analytics. But what does it really take to make it work in your stack? Instead of relying on a single database to do everything, this approach combines two mature open-source systems, each optimized for its specific purpose. It's the foundation behind real-time data platforms at companies like GitLab, Cloudflare, Instacart and thousands of others. You'll learn how to deploy Postgres and ClickHouse in your environment, set up real-time replication using PeerDB, and use the pg_clickhouse extension to query ClickHouse directly from Postgres — enabling a unified query layer for your applications.

​Key Takeaways

• ​Understand the Postgres + ClickHouse architectural pattern and when to apply it

• ​Learn how to set up real-time CDC replication using open source PeerDB

• ​Master the pg_clickhouse extension for unified querying across both databases

• ​Get hands-on experience building an open-source real-time data stack

• ​Explore operational best practices for running Postgres and ClickHouse together

​Prerequisites

• ​Basic understanding of data engineering concepts

• ​Familiarity with SQL

• ​Familiarity with Postgres internals (helpful but not required)

​Required Materials

• ​Laptop with Docker Desktop or OrbStack installed

• ​Postgres and ClickHouse installations (instructions provided ahead of the workshop)