ClickHouse OLAP

Design and query ClickHouse for real-time analytics. Models MergeTree tables, writes fast aggregations and materialized views, tunes partitions and indexes, and connects from Node and Python clients.

This skill applies ClickHouse analytics patterns for event and metrics data. It models MergeTree and ReplacingMergeTree tables, designs sort keys and partitions, writes high-throughput aggregations and materialized views, tunes query performance, and wires up Node and Python clients for ingestion and reads.

When to use

Use when designing ClickHouse schemas, writing analytical aggregations or materialized views, tuning partitions and indexes, or querying ClickHouse for real-time metrics.

Examples

Model an events table

Design a MergeTree schema

Design a ClickHouse MergeTree table for product events with the right sort key and monthly partitioning for fast time-range queries

Materialized view

Pre-aggregate metrics

Create a ClickHouse materialized view that rolls up raw events into hourly active-user counts

Optimize a query

Speed up an aggregation

This ClickHouse GROUP BY scans the whole table. Rewrite it to use the sort key and add the right index