It’s been exactly one month since I published my work porting Greenplum to PostgreSQL 19 and isolating the minimal set of PostgreSQL core patches it needs to run. It all started as a way to extract Greenplum’s ORCA analytical query planner and turn PostgreSQL into a modern, vectorized, high-performance DBMS. And I pulled it off! After a month of development, the database keeps getting faster. I’m now working on a GPU version. Even as a rough draft, on my MSI Raider A18 HX A9W laptop it runs queries faster than LTS ClickHouse on hot runs (the best of three tries): on MergeTree, 1.9x faster by shifted geometric mean (ClickHouse: 67.4 ms) and 1.3x by total time (ClickHouse: 7.99 s), and on Parquet through vexec_parquet, 13x faster by shifted geometric mean (ClickHouse: 229.1 ms) and 18x by total time (ClickHouse: 21.23 s).

I’d like to share the results of pg_vexec, an experimental vectorized extension for PostgreSQL whose CPU version beats ClickHouse on query execution time on Parquet and nearly catches up on MergeTree. I’ve published the telemetry from its ClickBench runs. It should be an interesting read, with plenty of charts, metrics, DBMS logs and numbers from deep inside the engine: https://igor-suhorukov.github.io/sql_report/

What pg_vexec is: https://igor-suhorukov.github.io/blog/2026/10/07/vectorized-postgresql-extension/

Source code for the CPU version: