Re: Question about nested loops..
| От | Andrei Lepikhov |
|---|---|
| Тема | Re: Question about nested loops.. |
| Дата | |
| Msg-id | d10b91e6-76eb-4a6c-bb17-c5736577f5ea@gmail.com обсуждение исходный текст |
| Ответ на | Question about nested loops.. (Frits Jalvingh <jal@etc.to>) |
| Список | pgsql-performance |
On 10/10/2025 10:51, Frits Jalvingh wrote:
> Hi Andrei,
> Thanks a lot for your response. I do not fully get it though because the
> well-performing query also references the outer (external?) relation:
> -> where eenheid.id_h_eenheid = eenheid_s.id_h_eenheid
> But your response made me think a but more.This is the schema of EXPLAIN for the second 'optimal' query:
Seq Scan on s_h_eenheid_ssm D
SubPlan 2
-> Hash Join
Hash Cond: (A.x = B.x)
-> Seq Scan on A
Filter: ...
-> Hash
-> Hash Join
Hash Cond: (C.y = B.x)
-> Seq Scan on s_h_huurovereenkomst_ssm C
-> Hash
-> Seq Scan on s_h_eenheid_ssm B
There is no reference to the upper relation D - each HashJoin refers to
its left and right side only.
-- regards, Andrei Lepikhov, pgEdge
В списке pgsql-performance по дате отправления: