ClickHouse
By ClickHouse, Inc.
ClickHouse is an open-source, column-oriented database management system built for online analytical processing (OLAP), capable of executing aggregation and analytical queries over billions of rows in sub-second time.
Definition
ClickHouse is an open-source, column-oriented database management system built for online analytical processing (OLAP), capable of executing aggregation and analytical queries over billions of rows in sub-second time.
Overview
Most transactional databases like PostgreSQL or MySQL store data row by row, which is efficient for reading and writing individual records but slow for queries that scan and aggregate huge numbers of rows — exactly the kind of workload analytics dashboards and reporting systems generate. ClickHouse instead stores data column by column: each column is compressed and read independently, so a query that only needs a handful of columns out of a hundred-column table can skip the rest entirely, dramatically reducing I/O. On top of columnar storage, ClickHouse uses vectorized query execution — processing data in batches using CPU-friendly operations — along with aggressive compression and a variety of specialized table engines (including the widely used MergeTree family) tuned for different ingestion and query patterns. This combination lets a single ClickHouse server routinely scan and aggregate billions of rows per second. Originally developed at Yandex and open-sourced in 2016, ClickHouse is now maintained by ClickHouse, Inc. and used heavily for real-time analytics, log and event analysis, observability backends, and business intelligence, often sitting alongside data warehouses like Snowflake or BigQuery, and streaming systems like Kafka that feed it live data. Its SQL interface and speed make it a natural fit for hands-on practice in a SQL Mastery course.
Key Features
- Column-oriented storage optimized for analytical (OLAP) query patterns
- Vectorized query execution for high-throughput CPU-efficient processing
- MergeTree family of table engines tuned for large-scale ingestion
- Real-time query performance over billions of rows
- Native SQL interface with extensions for arrays, nested data, and approximate functions
- Built-in data compression that reduces storage footprint significantly
- Distributed, sharded deployments for horizontal scale
- Integrations with Kafka, S3, and common BI tools