Skip site navigation (1) Skip section navigation (2)

Re: Multiprocessor performance

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "Valentin Puente" <vpuente(at)atc(dot)unican(dot)es>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Multiprocessor performance
Date: 2001-06-05 14:04:56
Message-ID: 15142.991749896@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-hackers
"Valentin Puente" <vpuente(at)atc(dot)unican(dot)es> writes:
> Ok...well I'm running this benchmarks in different SMP machines (SGI with 4
> to 8 processors and the results are odd). The best performance is achieved
> with just one backend (1 client). When I try to run more clients the tps
> falls quickly.

What scale factor (-s parameter for pgbench init) are you using for the
benchmark?

At scale factor 1, there's only one "branch" row, so all the
transactions have to update the same row and naturally will spend most
of their time waiting to do so.

You want scale factor >> # of concurrent clients to avoid interlock
effects.

			regards, tom lane

In response to

pgsql-hackers by date

Next:From: Oleg BartunovDate: 2001-06-05 14:07:47
Subject: Strange query plan
Previous:From: Vince VielhaberDate: 2001-06-05 13:55:49
Subject: Re: Mirrors not tracking main ftp site?

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group