How VK Inserts Data into ClickHouse from Tens of Thousands of Servers
summary
VK's use of ClickHouse for storing logs and statistics at large scale, including how much data is kept and how it is written from tens of thousands of servers. The talk covers ClickHouse performance under VK's workload, cluster configurations, and the use of buffer tables to ingest data.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 6:32Why ClickHouse for VK
- 13:40Direct insert problems
- 22:47Buffer tables drawbacks
- 28:44Data reader tool
- 32:17Summary and ClickHouse tips
related talks
48mHow VK Inserts Data into ClickHouse from Tens of Thousands of Servers
26mHandling Large Volumes of Immutable Structured Data With ClickHouse
51mWhat You Need to Know About ClickHouse Architecture
42mClickHouse: what is behind the fastest columnar database
31mHow ClickHouse® Inspired Us to Build a High Performance Time Series Database
59mEffective Use of ClickHouse
49mSolving real business cases with ClickHouse
34mClickHouse® - What is Behind the Fastest Columnar Database?