Re: Avoiding hash join batch explosions with extreme skew and weird stats - Mailing list pgsql-hackers

From Peter Geoghegan
Subject Re: Avoiding hash join batch explosions with extreme skew and weird stats
Date
Msg-id CAH2-Wz=JPP02V-CxUvDqVkN=wWtvGfoHDTrU5x5Jsbujk-b6oA@mail.gmail.com
Whole thread Raw
In response to Re: Avoiding hash join batch explosions with extreme skew and weird stats  (Melanie Plageman <melanieplageman@gmail.com>)
List pgsql-hackers
On Tue, Jul 30, 2019 at 8:07 PM Melanie Plageman
<melanieplageman@gmail.com> wrote:
> For the actual write to disk, I'm pretty sure I get that for free from
> the BufFile API, no?
> I was more thinking about optimizing when I call BufFileWrite at all.

Right. Clearly several existing buffile.c users regularly have very
small BufFileWrite() size arguments. tuplestore.c, for one.

-- 
Peter Geoghegan



pgsql-hackers by date:

Previous
From: Melanie Plageman
Date:
Subject: Re: Avoiding hash join batch explosions with extreme skew and weird stats
Next
From: Kohei KaiGai
Date:
Subject: How to retain lesser paths at add_path()?