Columnstore Revolution
summary
ClickHouse claims to handle billions of daily clickstream rows and delivers two orders of magnitude faster analytic scans than legacy MySQL, powering observability at huge fleets and telemetry-heavy users like Tesla; ClickHouse combines LSM-style ingestion with columnar storage and works with Parquet/Iceberg lakes. DuckDB is a tiny embeddable engine that runs on a laptop or inside Python, used on satellites to compress telemetry to Parquet, and excels as a local sandbox and library. AI agents drive far higher query volumes and much of coding and review becomes AI-augmented.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:57ClickHouse origins
- 6:32DuckDB design
- 8:38ClickHouse vs DuckDB
- 10:03ClickHouse analytics
- 12:28DuckDB embedded
- 14:17Where ClickHouse outperforms DuckDB
- 16:00Open formats support
- 19:15AI in databases
- 23:45AI in development
- 27:39Closing remarks
related talks
42mClickHouse: what is behind the fastest columnar database
45mClickHouse: what is behind the fastest columnar database
34mClickHouse® - What is Behind the Fastest Columnar Database?
26mColumnar Storage: Redefining Data Management for the Modern Era
40mData warehouse or conventional database: Which is right for you?
30mSQL Row Store vs Data Warehouse: Which Is Right for Your Application?
17mBuilding Real-Time Data Lake with ClickHouse
51mWhat You Need to Know About ClickHouse Architecture