DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=message-id:date:from:to:subject:cc:in-reply-to:mime-version
	:content-type:content-transfer-encoding:content-disposition
	:references;
	b=AjeObZPcAn/uCFTPv6Nh+2hXHc5jPB3HTad7TY1g36Y3MQtMVsFvG43yqli7X+R6ky
	zVEkqZ1KC/kv0b7xFngBBUcEtzb+UI5Z6kqysw64BEXFqH6w6bq0E6Ld/HRoOUX4b+MG
	l0Gtvtq/jE/oP9pmiE4tVKMnWfOoA2wDt4MLg=
Message-ID: <dcc563d10811241324t116f6e71j4863632b2c0c09a5@mail.gmail.com>
Date: Mon, 24 Nov 2008 14:24:47 -0700
From: "Scott Marlowe" <scott.marlowe@gmail.com>
To: "Brad Nicholson" <bnichols@ca.afilias.info>
Subject: Re: Monitoring buffercache...
Cc: "Kevin Kempter" <kevin@consistentstate.com>,
	pgsql-performance@postgresql.org
In-Reply-To: <1227556326.18990.98.camel@bnicholson-desktop>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <200811241143.57182.kevin@kevinkempterllc.com>
	<dcc563d10811241146l95302b1wd0e88756883e4401@mail.gmail.com>
	<1227556326.18990.98.camel@bnicholson-desktop>

On Mon, Nov 24, 2008 at 12:52 PM, Brad Nicholson
<bnichols@ca.afilias.info> wrote:
>> I just ran it in a loop over and over on my 8 core opteron server and
>> it ran the load factor up by almost exactly 1.0.  Under our normal
>> daily load, it sits at 1.9 to 2.5, and it climbed to 2.9 under the new
>> load of running that query over and over.  So, it doesn't seem to be
>> blocking or anything.
>
> The internal docs for pg_buffercache_pages.c state:
>
> "To get a consistent picture of the buffer state, we must lock all
> partitions of the buffer map.  Needless to say, this is horrible
> for concurrency.  Must grab locks in increasing order to avoid
> possible deadlocks."

Well, the pg hackers tend to take a parnoid view (it's a good thing
TM) on things like this.  My guess is that the period of time for
which pg_buffercache takes locks on the buffer map are short enough
that it isn't a real big deal on a fast enough server.  On mine, it
certainly had no real negative effects for the 5 minutes or so it was
running in a loop.  None I could see, and we run hundreds of queries
per second on our system.

Of course, for certain other types of loads it could be a much bigger
issue.  But for our load, on our machine, it was virtually
unnoticeable.