Re: How to get around LIKE inefficiencies? - Mailing list pgsql-hackers

From Tatsuo Ishii
Subject Re: How to get around LIKE inefficiencies?
Date
Msg-id 20001107101141R.t-ishii@sra.co.jp
Whole thread Raw
In response to How to get around LIKE inefficiencies?  (The Hermit Hacker <scrappy@hub.org>)
List pgsql-hackers
>     I'm running this on a Dual-PIII 450 Server, 512Meg of RAM, zero
> swap space being used ... the database has its indices on one hard drive,
> the tables themselves are on a second one ... its PgSQL 7.0.2 (Tom,
> anything in v7.0.3 that might improve this?) and startup is as:
> 
> #!/bin/tcsh
> setenv PORT 5432
> setenv POSTMASTER /pgsql/bin/postmaster
> unlimit
> ${POSTMASTER} -B 384 -N 192 \
>               -o "-F -S 32768" \
>               -i -p ${PORT} -D/pgsql/data >&
> /pgsql/logs/postmaster.${PORT}.$$ &

BTW, you have a 32MB sort space for each backend, and you allow up to
192 concurrent backends. So whole postgres requires at least 192*32MB
= 6144MB memory if all 192 users try to connect to your server at the
same time!  I would suggest adding enough swap space or descreasing -S
setting...
--
Tatsuo Ishii


pgsql-hackers by date:

Previous
From: Tatsuo Ishii
Date:
Subject: Re: RE: [COMMITTERS] pgsql/src/backend/access/transam (xact.c xlog.c)
Next
From: "Dominic J. Eidson"
Date:
Subject: Re: Darn, long option emulation doesn't work on FreeBSD