ClickHouse Documentation

What ClickHouse is

ClickHouse is an open source, column-oriented database for online analytical processing (OLAP). Instead of storing each row together, it stores each column together and compresses it, so a query like "total sales per city for the last 90 days" reads only the three columns it needs. That makes aggregations over hundreds of millions of rows fast on modest hardware. It speaks SQL, with its own extensions.

What the docs cover

• Getting started: installing with one command (curl https://clickhouse.com/ | sh) on Linux or macOS, clickhouse local for querying files without a server, and ClickHouse Cloud.
• Table engines: the MergeTree family, sorting keys and why the order of columns in the key matters for speed.
• SQL reference: functions, aggregate functions, joins and data types.
• Ingesting data: from files, Kafka, S3 and other databases.
• Use-case guides: observability, real-time analytics dashboards and data warehousing.

Is it worth learning?

If you are building analytics into a product (dashboards over events, click tracking, logs or IoT readings) and your main database is getting slow at it, ClickHouse is one of the most common answers. It also teaches you what OLTP versus OLAP really means, because things that are cheap in PostgreSQL (updating one row) are expensive here, and things that are slow in PostgreSQL (scanning a billion rows) are quick.

When not to use it

Do not make it the main database for users, orders or payments. It is built for many reads over large data, not for lots of small transactional updates. Most apps keep PostgreSQL or MySQL as the source of truth and copy events into ClickHouse for analysis.

Prerequisites and time

Solid SQL basics. A weekend to load a public dataset, try different sorting keys and compare query times.

Free or paid?

The open source server is free under Apache 2.0. ClickHouse Cloud is a paid managed service with a free trial for new accounts.

Best for: backend and data engineers adding fast analytics to a product, and anyone curious how columnar databases work.