From: | Herwig Goemans <herwig(dot)goemans(at)gmail(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-sql(at)postgresql(dot)org |
Subject: | Re: why is PG NOT pushing where clause to VIEW ? Resulting in bad performance. |
Date: | 2017-02-08 10:20:14 |
Message-ID: | CAKaDz_GHq5KTGQ4iFS5g2DrNVZkaBdFZj=eQ-mmWocOuBbQZ_A@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
Hello,
I do not see how pushing this clause to to the view can alter in any way
the result. If the latter is true, it is obvious to implement this. A short
test without data in another database showed no change in plan with or
without the extra clause, in fact it said access predicates on currency
view on ID and on wsid. So it looks to me this is a opportunity for PG to
add these predicates. The benefit performante wise is huge: the query goes
down from 4-6s elapsed time to less than 1s.
Kind regards and thanks for the help.
Herwig
Op 7 feb. 2017 5:37 p.m. schreef "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
Herwig Goemans <herwig(dot)goemans(at)gmail(dot)com> writes:
> I am wondering why a certain query is NOT pushing the where clause to the
> view. See query and explain plan:
There's no provision for transitively applying IN clauses to other
variables. I'm not entirely sure whether it would be a good idea,
but in any case, it's not there and not planned.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Mike Wes | 2017-02-08 11:20:35 | Problem wil TIMEZONE vs TIME ZONE |
Previous Message | Tom Lane | 2017-02-07 16:37:33 | Re: why is PG NOT pushing where clause to VIEW ? Resulting in bad performance. |