Using ClickHouse® Database to Power an Analytics and Customer Engagement Platform
summary
Times Internet describes GrowthRx, an in‑house analytics and engagement platform built on ClickHouse that ingests 30–40k req/s (tested to 100k by adding nodes), batches via Kafka consumers into ClickHouse on three VM replicas with 10 TB SSDs, and produces real‑time segments of 10–50+ million users. Segmentation uses bitmap functions processing >200M rows/sec; dashboards rely on materialized views, sample clause (e.g., 10% at ~1B rows) and parametric/window funnel functions, with async queries written to object storage to avoid CPU contention.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:37About Times Internet
- 3:30Architecture
- 4:35Data ingestion
- 5:42Segmentation
- 7:35Dashboarding
- 9:05Key Features
- 9:22Metallized View
- 10:56Sample Clause
- 12:38Funnel View
- 13:48Questions
related talks
42mClickHouse: what is behind the fastest columnar database
10mRealtime DNS Analytics at Cloudflare with #ClickHouse
55mBuilding Fast ClickHouse Analytics for MySQL Transaction Data Using Kubernetes
30mBuilding For Fast
28mColumnstore Revolution
26mHandling Large Volumes of Immutable Structured Data With ClickHouse
27mFast DIY O11y with OpenSrc Analytics & Visualization
22mIt’s all about the db - a journey to accommodate 100k events per second