Subject: Re: pg 8.1.3, AIX, huge box, painfully slow.
From: Simon Riggs <simon@2ndquadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Luke Lonergan <llonergan@greenplum.com>,
	Gavin Hamill <gdh@laterooms.com>, pgsql-performance@postgresql.org
In-Reply-To: <17613.1144451138@sss.pgh.pa.us>
References: <C05C3C44.211A5%llonergan@greenplum.com>
	<17613.1144451138@sss.pgh.pa.us>
Content-Type: text/plain
Date: Mon, 10 Apr 2006 10:26:49 +0100
Message-Id: <1144661209.21409.131.camel@localhost.localdomain>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit

On Fri, 2006-04-07 at 19:05 -0400, Tom Lane wrote:

> It's plausible though that we are seeing contention across members of
> the LWLock array, with the semop storm just being a higher-level symptom
> of the real hardware-level problem.  You might try increasing
> LWLOCK_PADDED_SIZE to 64 or even 128, see
> src/backend/storage/lmgr/lwlock.c (this is something that does exist in
> 8.1, so it'd be easy to try).

pSeries cache lines are 128 bytes wide, so I'd go straight to 128.

If you're renting all 8 CPUs, I'd drop to 4 and try that instead. With 8
CPUs the contention will vary according to what each CPU is doing at any
one time - when they all hit the contention spot, things will get worse.

The pSeries has good CPUs and great caching, so I'd expect contention to
be somewhat more apparent as a bottleneck.

Best Regards, Simon Riggs