From owner-freebsd-hardware@FreeBSD.ORG  Fri Oct 31 09:11:07 2008
Return-Path: <owner-freebsd-hardware@FreeBSD.ORG>
Delivered-To: freebsd-hardware@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 066271065686
	for <freebsd-hardware@freebsd.org>;
	Fri, 31 Oct 2008 09:11:07 +0000 (UTC)
	(envelope-from numisemis@yahoo.com)
Received: from web36607.mail.mud.yahoo.com (web36607.mail.mud.yahoo.com
	[209.191.85.24]) by mx1.freebsd.org (Postfix) with SMTP id A57A58FC28
	for <freebsd-hardware@freebsd.org>;
	Fri, 31 Oct 2008 09:11:06 +0000 (UTC)
	(envelope-from numisemis@yahoo.com)
Received: (qmail 73161 invoked by uid 60001); 31 Oct 2008 08:44:25 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:X-Mailer:Date:From:Reply-To:Subject:To:Cc:MIME-Version:Content-Type:Message-ID;
	b=ZWXbW9ZIC3mWOaV4MWsfDFKPAsc/LAp8Uq0GmN8veYBj/1FIspiNW1d+m22XKeFgl0so0oguTnJxgKJvwD0DrfGU618d9pN4FyuntcDlJLPgyJsOJ9n2n8BJCX/NkJd6p7Z93MxcrqaaVjvQR5mTLHD+htA7dARtZDDGgNiY2Wg=;
X-YMail-OSG: bm_9wYwVM1mhYQn4Mm3qK.xkJubVu_xhfEAjZuexXYGOTy6Z_nnvYq.bcRIAxGvxdwOPwz164XyEKGx7uFd98wO04Eicwi9Fvqi66xSAhxNlLRWXIcmON6mQUfoA09NF5BpYvfOBiQnbbbr3j1Ebasb0Kz1VlHR47yjaLRmE
Received: from [213.147.110.159] by web36607.mail.mud.yahoo.com via HTTP;
	Fri, 31 Oct 2008 01:44:25 PDT
X-Mailer: YahooMailWebService/0.7.260.1
Date: Fri, 31 Oct 2008 01:44:25 -0700 (PDT)
From: Simun Mikecin <numisemis@yahoo.com>
To: Jeremy Chadwick <koitsu@FreeBSD.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Message-ID: <580369.72525.qm@web36607.mail.mud.yahoo.com>
X-Mailman-Approved-At: Fri, 31 Oct 2008 11:24:49 +0000
Cc: Danny Carroll <fbsd@dannysplace.net>, reebsd-fs@freebsd.org,
	freebsd-hardware@freebsd.org
Subject: Re: Areca vs. ZFS performance testing.
X-BeenThere: freebsd-hardware@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: numisemis@yahoo.com
List-Id: General discussion of FreeBSD hardware <freebsd-hardware.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-hardware>, 
	<mailto:freebsd-hardware-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-hardware>
List-Post: <mailto:freebsd-hardware@freebsd.org>
List-Help: <mailto:freebsd-hardware-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-hardware>, 
	<mailto:freebsd-hardware-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Fri, 31 Oct 2008 09:11:07 -0000

Jeremy Chadwick wrote:
> The tuning variables I advocate for a system with 2GB of RAM or more,
> on RELENG_7, are:
> vm.kmem_size="1536M"
> vm.kmem_size_max="1536M"

There is no point in setting vm.kmem_size_max. Setting vm.kmem_size is enough. vm.kmem_size_max is used for auto-tuning of kmem size which is in this case actually overriden by manually setting vm.kmem_size.

> vfs.zfs.arc_min="16M"
> vfs.zfs.arc_max="64M"
> vfs.zfs.prefetch_disable="1"
> You can gradually increase arc_min and arc_max by ~16MB increments as
> you see fit; you should see general performance improvements as they
> get larger (more data being kept in the ARC), but don't get too crazy.
> I've tuned arc_max up to 128MB before with success, but I don't want
> to try anything larger without decreasing kmem_size_*.

Can you explain why would you have to decrease kmem_size to use larger ARC?
AFAIK it should be contrary to what you are saying: when you use larger kmem_size you can also use larger arc_max.

My suggestion if you are using kmem_size of 1536M would be to not tune arc_min and arc_max if your system isn't panicing. If it does you should try decreasing arc_max (from it's default value) until it doesn't.