From owner-freebsd-questions@FreeBSD.ORG Fri Sep 16 21:22:48 2011 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CACAF106566C for ; Fri, 16 Sep 2011 21:22:48 +0000 (UTC) (envelope-from terje@elde.net) Received: from keepquiet.net (keepquiet.net [IPv6:2a01:4f8:130:84c1::deaf:babe]) by mx1.freebsd.org (Postfix) with ESMTP id 8B0488FC0C for ; Fri, 16 Sep 2011 21:22:48 +0000 (UTC) Received: from [10.0.0.30] (13.80-203-88.nextgentel.com [80.203.88.13]) (using TLSv1 with cipher ECDHE-RSA-AES128-SHA (128/128 bits)) (Client did not present a certificate) (Authenticated sender: terje@elde.net) by keepquiet.net (Postfix) with ESMTPSA id A7FDC2E1C8; Fri, 16 Sep 2011 23:22:43 +0200 (CEST) References: <20110916063153.200375qdq59crf8c@mail.top-consulting.net> <32990703-D068-4B0D-AF3A-C1E6EA0A4100@elde.net> <20110916101833.17485ybnq5srjbc4@mail.top-consulting.net> In-Reply-To: <20110916101833.17485ybnq5srjbc4@mail.top-consulting.net> Mime-Version: 1.0 (iPhone Mail 8C148) Content-Type: text/plain; charset=us-ascii Message-Id: Content-Transfer-Encoding: quoted-printable X-Mailer: iPhone Mail (8C148) From: Terje Elde Date: Fri, 16 Sep 2011 23:22:30 +0200 To: "freebsd@top-consulting.net" Cc: "freebsd-questions@freebsd.org" Subject: Re: FS of choice for max random iops ( Maildir ) X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 16 Sep 2011 21:22:48 -0000 On 16. sep. 2011, at 16:18, freebsd@top-consulting.net wrote: > Got a measly 74MB/sec. You can't ask for advice, get it, do something completely different, and the= n complain that it didn't work.=20 Neither can you ask people to donate their time, if you won't spend yours.=20= In other words: if you won't listen, there's no point in us talking.=20 However: Don't disable ZIL. Just don't. It's not the way to go. If you want to know w= hy, google will help.=20 Also, you're making some assumptions, such as the ZIL being bad for performa= nce. That's not always the case. ZIL-writes are a rather nice load for spinn= ing metal storage. Even if you write through cache, that can give you a boos= t on your real world workload.=20 Which brings us to the third bit. You're benchmarking, not trying real world= loads. That's the load you'll have to worry about, and it's the load zfs sh= ines at.=20 Thanks to the ZIL (the thing you're trying to kill, remember?) you can conve= rt seek heavy writes to sequential zil-writes, freeing up disk bandwith for c= oncurrent reads.=20 If you want to test before spending money, try what Svein said. Set up a sma= ll logical volume (preferrably smaller than your controller cache, if it's l= arge enough), then try that as a dedicated zil-device.=20 Never tried that, but worth a shot.=20 Terje=