On Wed, Jul 15, 2020 at 9:02 PM Etsuro Fujita <etsuro.fuj...@gmail.com> wrote: > On Wed, Jul 15, 2020 at 12:12 AM Alexey Kondratov > <a.kondra...@postgrespro.ru> wrote: > > On 2020-07-14 15:27, Ashutosh Bapat wrote: > > > On Tue, Jul 14, 2020 at 12:48 AM Alexey Kondratov > > > <a.kondra...@postgrespro.ru> wrote: > > >> Some real-life test queries show, that all single-node queries aren't > > >> pushed-down to the required node. For example: > > >> > > >> SELECT > > >> * > > >> FROM > > >> documents > > >> INNER JOIN users ON documents.user_id = users.id > > >> WHERE > > >> documents.company_id = 5 > > >> AND users.company_id = 5; > > > > > > There are a couple of things happening here > > > 1. the clauses on company_id in WHERE clause are causing partition > > > pruning. Partition-wise join is disabled with partition pruning before > > > PG13. > > More precisely, PWJ cannot be applied when there are no matched > partitions on the nullable side due to partition pruning before PG13.
On reflection, I think I was wrong: the limitation applies to PG13, even with advanced PWJ. > But the join is an inner join, so I think PWJ can still be applied for > the join. I think I was wrong in this point as well :-(. PWJ cannot be applied to the join due to the limitation of the PWJ matching logic. See the discussion started in [1]. I think the patch in [2] would address this issue as well, though the patch is under review. Best regards, Etsuro Fujita [1] https://www.postgresql.org/message-id/CAN_9JTzo_2F5dKLqXVtDX5V6dwqB0Xk%2BihstpKEt3a1LT6X78A%40mail.gmail.com [2] https://www.postgresql.org/message-id/502.1586032...@sss.pgh.pa.us
