[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgresql-general
Subject:    Re: SELECT with LIKE clause makes full table scan
From:       Matthias Apitz <guru () unixarea ! de>
Date:       2022-01-26 15:28:33
Message-ID: YfFoobLLLF5wjPFp () c720-r368166
[Download RAW message or body]

El día miércoles, enero 26, 2022 a las 11:21:12p. m. +0800, Julien Rouhaud escribió:

> Hi,
> 
> On Wed, Jan 26, 2022 at 11:07 PM Matthias Apitz <guru@unixarea.de> wrote:
> >
> > We changed two relevant Indexes to
> >
> > CREATE INDEX d01ort ON d01buch(d01ort bpchar_pattern_ops );
> > CREATE INDEX d01ort2 ON d01buch(d01ort2 bpchar_pattern_ops );
> 
> When you said changed, did you drop the previous ones?  

Yes, of course.

> As Tom
> mentioned, those indexes are specialized and are only useful for LIKE
> 'something%' queries.  It's quite likely that your existing indexes
> were useful for other queries, which may not be as fast without those
> indexes.  You can check in pg_stat_user_indexes if your indexes seems
> to be used before actually dropping them for instance:
> https://www.postgresql.org/docs/current/monitoring-stats.html#MONITORING-PG-STAT-ALL-INDEXES-VIEW

Thanks, we're still investigating more cases with LIKE clause and will
consider your hint.

	matthias


-- 
Matthias Apitz, ✉ guru@unixarea.de, http://www.unixarea.de/ +49-176-38902045
Public GnuPG key: http://www.unixarea.de/key.pub


[prev in list] [next in list] [prev in thread] [next in thread]