> ClickHouse performs what is known as PREWHERE processing > p.s. Scans are ridiculously fast. Good point, I should have mentioned this was basically a worst-c

> ClickHouse performs what is known as PREWHERE processing > p.s. Scans are ridiculously fast. Good point, I should have mentioned this was basically a worst-case scenario for Clickhouse as the data layout is totally random (same approach as OP used in their benchmark) and isn't able to utilize any granule pruning, sorting, or skip indexing, but is still able to achieve such remarkable speeds.