ClickHouse Splits PREWHERE into Read Steps: Bytes per Rejected Row Picks the Order
ClickHouse moves WHERE conditions into PREWHERE by cost per rejected row, and reads PREWHERE columns in separate read steps.
Posts
All posts published on this site.
Everything published here, newest first.
ClickHouse Flattens the Join Tree into a QueryGraph: A Greedy Search and a Full Plan Rebuild
ClickHouse flattens the join tree into a graph, searches it for a cheaper order, and rebuilds the plan.
ClickHouse Guesses Join Sizes from Distinct-Value Counts: One Over the Larger NDV
ClickHouse’s join cost model is built from NDV counts, min/max ranges and row estimates.
ClickHouse Searches Join Orders with Dynamic Programming: Three Algorithms Over One Memo Keyed by Bitsets
ClickHouse ships three DP join-order searches over a relation-set memo, with a plan budget and a fallback chain.
ClickHouse Reads a 5-Bit Rank Through an Unaligned 16-Bit Window: The CompactArray Trick
ClickHouse’s CompactArray stores sub-byte values in a bitset and reads them through an unaligned 16-bit window, so a HyperLogLog rank costs 5 or 6 bits per bucket.