From: Tim Uckun <tim@diligence.com>
To: pgsql-interfaces@postgresql.org
Subject: Access and Postgres question.
Date: Fri, 28 Jul 2000 20:49:53 -0600
Message-ID: <4.2.0.58.20000728204043.00c24d20@mail.diligence.com> (raw)
I am testing postgres 7.02 and ms-sql server as back ends to ms access. I
created identical databases containing three tables with about 1.2 , 1.6
and 3 million records each. The machine that is running SQL server is a
dual pIII zeon machine with raid 0 and 512 megs of ram. The postgres
machine is a dual pIII 500 with 256 megs of ram no raid but nice fast scsi
drive running red hat. Given this scenario I do expect postgres to be
slower then ms sql server. What I find very odd is that in some tasks it
actually faster (very slightly) like when doing simple select * from table
where primarykey=x. But when I do select * from very_big_table (I know it's
not such a good idea) it is insanely slow. On one test the table with 1.2
million records too 12 seconds in postgres and less then a second in sql
server!. I suspect that this has something to do with the ODBC driver
because I am having a hard time believing that postgres is that slow. I
suspect that the MS-SQL server ODBC driver is not actually pulling up ALL
the records but just a few at a time (and perhaps spinning off another
thread to fetch the rest). Can anybody confirm this?
----------------------------------------------
Tim Uckun
Mobile Intelligence Unit.
----------------------------------------------
"There are some who call me TIM?"
----------------------------------------------
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-interfaces@postgresql.org
Cc: tim@diligence.com
Subject: Re: Access and Postgres question.
In-Reply-To: <4.2.0.58.20000728204043.00c24d20@mail.diligence.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox