Building For Fast

on YouTube ↗30m

" summary

ClickHouse demonstrates concrete performance: a public GitHub-events query returns star counts in 14 milliseconds (about 60 such queries per two heartbeats), bulk ingest reaches 1.13 billion rows in 65 seconds (~17 million rows/sec), and a Monte Carlo run evaluates ~100 billion samples in 11 seconds. Engineering delivered a custom consensus service ClickHouseKeeper to replace ZooKeeper, maintains monthly releases, publishes open results via clickbench, and readies a managed service ClickHouse Cloud for early October.

AI-written summary · about

" chapters (14)

1. 0:00Introduction and welcome
2. 1:05Milliseconds matter everywhere
3. 5:43What is ClickHouse
4. 8:34Defining fast in databases
5. 10:21Fast operations with ClickHouse Cloud
6. 11:26Demo spinning up a service
7. 15:41Querying remote data with S3
8. 16:04Ingesting 1 billion rows
9. 18:55Running analytics queries
10. 25:13Non-technical aspects of fast
11. 25:56Monthly release cycle
12. 27:02ClickHouse Keeper story
13. 28:47ClickBench benchmark
14. 29:54Conclusion and Q&A

" related talks

• 42mClickHouse: what is behind the fastest columnar databaseOlena Kutsenko · Berlin Buzzwords 2023
• 34mClickHouse® - What is Behind the Fastest Columnar Database?Olena Kutsenko · OSA Con 2022
• 59mEffective Use of ClickHouseAleksey Milovidov · HighLoad++ 2017
• 28mColumnstore RevolutionPeter Zaitsev, Hannes Mühleisen, Alexy Milovidov · Percona Live Amsterdam 2026
• 17mBuilding Real-Time Data Lake with ClickHouseAlexander Zaitsev · PG Armenia x Percona University – Yerevan, August 28, 2025
• 04mRoad to Percona Live 2022 - Robert HodgesRobert Hodges · Percona Live 2022
• 55mBuilding Fast ClickHouse Analytics for MySQL Transaction Data Using KubernetesRobert Hodges, Kanthi Subramanian · Percona Live 2022
• 26mHandling Large Volumes of Immutable Structured Data With ClickHouseJim Hague · ACCU 2021 Online