RE: v7.1b4 bad performance

From: "Schmidt, Peter" <peter(dot)schmidt(at)prismedia(dot)com>
To: "'Bruce Momjian'" <pgman(at)candle(dot)pha(dot)pa(dot)us>, "Schmidt, Peter" <peter(dot)schmidt(at)prismedia(dot)com>
Cc: "'Tom Lane'" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "'Michael Ansley'" <Michael(dot)Ansley(at)intec-telecom-systems(dot)com>, "'pgsql-admin(at)postgresql(dot)org'" <pgsql-admin(at)postgresql(dot)org>
Subject: RE: v7.1b4 bad performance
Date: 2001-02-16 22:23:37
Message-ID: F1DC8388AD52D411B83B00D0B774D6EB1928DF@winmail.prismedia.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

> -----Original Message-----
> From: Bruce Momjian [mailto:pgman(at)candle(dot)pha(dot)pa(dot)us]
> Sent: Friday, February 16, 2001 10:59 AM
> To: Schmidt, Peter
> Cc: 'Tom Lane'; 'Michael Ansley'; 'pgsql-admin(at)postgresql(dot)org'
> Subject: Re: [ADMIN] v7.1b4 bad performance
>
>
> > > "Schmidt, Peter" <peter(dot)schmidt(at)prismedia(dot)com> writes:
> > > > However, what I'm interested in is the comparison between
> > > 7.0x and 7.1x.
> > > > Shouldn't the 7.1b4 database be faster? Have you (or anyone
> > > else) done this
> > > > comparison?
> > >
> > > 7.1 is considerably faster if you compare performance without -F.
> > > If you are comparing performance with -F, I'm not sure
> exactly how it
> > > stacks up.
> > >
> > > regards, tom lane
> > >
> >
> > Here are some of the results of my tests which are very
> consistent. I
> > dropped and recreated postgres db before each test so the
> only user tables
> > were the pgbench tables.
> >
> > 7.1b4(no -F) 49 tps
> > 7.03 (no -F) 64 tps
> > 7.03 (with -F) 121 tps
> >
> > The test machine is a Dell Dimension 800mz w/128 RAM and it
> is dedicated to
> > this testing. Pgbench is run on the same machine. Any ideas
> why my results
> > differ? I will continue testing and appreciate all comments.
> > Peter
>
> This is very strange. Is there any specific query that is causing the
> problem? Does EXPLAIN show the same output on both versions?
>

"pgbench" doesn't execute any complex sql...

Explain shows exactly the same output for both versions(7.03 & 7.1b4):

update accounts set abalance = abalance + 1 where aid = 1
Index Scan using accounts_pkey on accounts (cost=0.00..3.81 rows=1
width=30)

select abalance from accounts where aid = 1
Index Scan using accounts_pkey on accounts (cost=0.00..3.81 rows=1 width=4)

update tellers set tbalance = tbalance + 1 where tid = 1
Seq Scan on tellers (cost=0.00..1.12 rows=1 width=30)

update branches set bbalance = bbalance + 1 where bid = 1
Seq Scan on branches (cost=0.00..1.01 rows=1 width=26)

insert into history(tid,bid,aid,delta,time) values(1,1,1,1,'now')
Result (cost=0.00..0.00 rows=0 width=0)

Peter

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message Bruce Momjian 2001-02-16 22:26:26 Re: v7.1b4 bad performance
Previous Message Jie Liang 2001-02-16 21:50:48 Re: Re: [GENERAL] what means "INSERT xxx yyy" ?