To: Wei Weng <wweng@kencast.com>
Cc: Richard Huxton <dev@archonet.com>, pgsql-sql@postgresql.org
Subject: Re: concurrent connections is worse than serialization? 
In-reply-to: <1029338971.21470.9.camel@Monet> 
References: <1029271170.16828.7.camel@Monet>
	<200208141018.38875.dev@archonet.com>
	<1029334827.21470.2.camel@Monet>
	<200208141549.57646.dev@archonet.com>
	<1029338971.21470.9.camel@Monet>
Comments: In-reply-to Wei Weng <wweng@kencast.com>
	message dated "14 Aug 2002 11:29:26 -0400"
Date: Wed, 14 Aug 2002 12:48:31 -0400
Message-ID: <9256.1029343711@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Wei Weng <wweng@kencast.com> writes:
> On Wed, 2002-08-14 at 10:49, Richard Huxton wrote:
>> If it's not CPU, is the system going into swap or are you seeing a
>> lot of disk activity?

> I did hear a lot of disk noise when I ran the test. How do I tell if the
> "system is going into swap"?

Try running "vmstat 1" while doing the test.  If the "swap" columns show
any activity then you're swapping.  Watching free memory with "top" may
also be insightful.

I tend to agree with what I think Richard suspects: you haven't got
enough RAM for thirty concurrent server processes, and so the system is
wasting a lot of time swapping processes to disk.  What are the hardware
parameters, anyway?

			regards, tom lane