Getting started with ClickHouse v26.3

Connect to your ClickHouse instance, explore EDB-specific capabilities, and find resources for going further.

Connecting to ClickHouse

Once ClickHouse is installed and running, connect using clickhouse-client:

clickhouse-client

Verify you're running an EDB-packaged build:

SELECT version();

The output includes the EDB release, for example 26.3.37.3+1edb1.

Integrations

• Connecting to WarehousePG: read from and write to WarehousePG tables using the PostgreSQL table engine
• Reading Apache Iceberg® tables: query Iceberg tables stored in Amazon S3 or Azure Blob Storage
• Connecting AI agents with MCP: let AI agents and LLM-based tools query ClickHouse using the MCP server
• Monitoring Postgres with pg_stat_ch: capture per-query execution telemetry from Postgres and analyze it in ClickHouse

See also

For SQL reference, configuration, and operational guidance beyond what's covered here, refer to the upstream ClickHouse documentation:

• Creating tables
• Inserting data
• Writing queries
• SQL reference
• Table engines
• Server configuration
• Best practices