From owner-freebsd-stable@FreeBSD.ORG Wed Mar 9 15:40:59 2011 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3D44F106564A for ; Wed, 9 Mar 2011 15:40:59 +0000 (UTC) (envelope-from hausen@punkt.de) Received: from kagate.punkt.de (kagate.punkt.de [217.29.33.131]) by mx1.freebsd.org (Postfix) with ESMTP id C32E38FC21 for ; Wed, 9 Mar 2011 15:40:58 +0000 (UTC) Received: from hugo10.ka.punkt.de ([217.29.45.10]) by gate1.intern.punkt.de with ESMTP id p29Fevw6034522; Wed, 9 Mar 2011 16:40:57 +0100 (CET) Received: from [217.29.45.147] ([217.29.45.147]) by hugo10.ka.punkt.de (8.14.2/8.14.2) with ESMTP id p29FeuBT072537; Wed, 9 Mar 2011 16:40:56 +0100 (CET) (envelope-from hausen@punkt.de) Mime-Version: 1.0 (Apple Message framework v1082) Content-Type: text/plain; charset=iso-8859-1 From: "Patrick M. Hausen" In-Reply-To: Date: Wed, 9 Mar 2011 16:40:51 +0100 Content-Transfer-Encoding: quoted-printable Message-Id: References: <20110308114810.GA37554@icarus.home.lan> <4D775CF1.1010501@gmx.de> <20110309125132.GB62338@icarus.home.lan> To: Tom Evans X-Mailer: Apple Mail (2.1082) Cc: Matthias Andree , freebsd-stable@freebsd.org, Jeremy Chadwick Subject: Re: ZFS performance as the FS fills up? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 09 Mar 2011 15:40:59 -0000 Hello, Am 09.03.2011 um 14:28 schrieb Tom Evans: > I don't think this is accurate. Prefetch was certainly disabled by > default on 8.0 if you had 4GB of RAM or less, requiring the sysctl > vfs.zfs.prefetch_disable=3D0 to be set if you wanted prefetch and had > 4GB of RAM or less. My configuration is this: JBOD with 24 disks before the performance increase, now 30. 5 vdevs of 6 disks each, raidz2. 16 GB RAM, 8.2 amd64. vm.kmem_size=3D"12288M" vfs.zfs.arc_max=3D"8192M" vfs.zfs.prefetch_disable=3D"1" vfs.zfs.txg.timeout=3D"5" ------- net.inet.tcp.sendbuf_max=3D16777216 net.inet.tcp.recvbuf_max=3D16777216 net.inet.tcp.sendspace=3D65536 net.inet.tcp.recvspace=3D131072 kern.maxvnodes=3D250000 vfs.zfs.txg.write_limit_override=3D1073741824 After I added the last 6 disks net write performance (that means how quick Amanda fills the current vtape) at least doubled. Without any further Amanda tuning. I am aware that there are many components involved that affect performance, so what would be the best way to measure the ZFS part? Working crystal balls have become rare these days ;-) Kind regards, Patrick --=20 punkt.de GmbH * Kaiserallee 13a * 76133 Karlsruhe Tel. 0721 9109 0 * Fax 0721 9109 100 info@punkt.de http://www.punkt.de Gf: J=FCrgen Egeling AG Mannheim 108285