From: | Lincoln Yeoh <lyeoh(at)pop(dot)jaring(dot)my> |
---|---|
To: | <jeff(at)dgjc(dot)org>, <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: PostgreSQL transaction locking problem |
Date: | 2002-02-04 04:41:10 |
Message-ID: | 3.0.5.32.20020204124110.008c11c0@192.228.128.13 |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
At 08:51 PM 02-02-2002 -0800, Jeff Martin wrote:
>independent from the database server. Thus in a multi-user situation I
>gain. However, can I tune postgres to run several server processes as well?
>Can postgres run concurrent server processes in both persistant and
>non-persistant connection situations? I am beginning to think I should not
>have encapsulated my business logic in pg/sql but in PHP functions instead.
The persistent and nonpersistent stuff is done by PHP not by postgresql.
You just have to configure Postgresql to support the number of concurrent
backends you need.
The potential performance problem in your case is it's not as easy to scale
up the postgresql box - you are still limited to running it on one server
AFAIK. So if more of your processing is done by postgresql you'd hit the
limit sooner. Whereas if the PHP scripts bore the brunt of the CPU load,
then you could probably add cheap machines for the PHP stuff. That said has
anyone got postgresql running on an IBM mainframe :)? Pgbench figures would
be interesting...
Cheerio,
Link.
From | Date | Subject | |
---|---|---|---|
Next Message | Lincoln Yeoh | 2002-02-04 04:59:38 | Re: Distributing index's/tables/logs/etc. |
Previous Message | Jeff Davis | 2002-02-04 03:50:14 | Re: Comparision of DBs (new to list) |