> The upside is Postgres doesn't have to worry about page splits, so things like bulk inserts can be much more efficient. Not in the heap, but if you have any

> The upside is Postgres doesn't have to worry about page splits, so things like bulk inserts can be much more efficient. Not in the heap, but if you have any index on the table (I know, don’t do that for bulk loads, but many don’t / it isn’t feasible sometimes) then you’re still dealing with a B+tree (probably). Also, MySQL still gets the nod for pure bulk load speed via MySQLShell’s Parallel Import Utility [0]. You can of course replicate this in Postgres by manually splitting the input file and running multiple \COPY commands, but having a tool do it all in one is lovely. [0]: https://dev.mysql.com/doc/mysql-shell/8.0/en/mysql-shell-utilities-parallel-table.html https://dev.mysql.com/doc/mysql-shell/8.0/en/mysql-shell-uti...