Exactly-once held through nine failures. The window counts did not.
Kafka, Flink, ClickHouse and Iceberg stayed exact through nine failures, yet 10-second window counts fell to 59.8% of on-time events. Why, and the fixes.
13 min read
Topic
ClickHouse
Column-store analytics: MergeTree design, partition and ordering keys, materialized views, and keeping system log tables from quietly eating the disk.
I have worked with ClickHouse since 2021, including in production systems and independent product experiments. It rewards careful ordering keys, partitions, and read-shaped models, then punishes weak choices quietly through merge pressure, part counts, and disk growth rather than a clean error. These notes focus on reproducible mechanics: MergeTree design, materialized views as incremental aggregation, query evidence, and the maintenance that keeps system logs from becoming the largest workload.