From owner-freebsd-questions@FreeBSD.ORG Fri Sep 16 21:39:59 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 B564B106566B for ; Fri, 16 Sep 2011 21:39:59 +0000 (UTC) (envelope-from freebsd@top-consulting.net) Received: from smtp.emailarray.com (mail.emailarray.com [69.28.212.222]) by mx1.freebsd.org (Postfix) with ESMTP id 4AE6E8FC17 for ; Fri, 16 Sep 2011 21:39:59 +0000 (UTC) Received: (qmail 70673 invoked by uid 89); 16 Sep 2011 21:39:58 -0000 Received: from unknown (HELO localhost) (freebsd@top-consulting.net@192.168.1.6) (POLARISLOCAL) by smtp.emailarray.com with (DHE-RSA-AES256-SHA encrypted) SMTP; 16 Sep 2011 21:39:57 -0000 Received: from modemcable238.94-200-24.mc.videotron.ca (modemcable238.94-200-24.mc.videotron.ca [24.200.94.238]) by mail.top-consulting.net (Horde Framework) with HTTP; Fri, 16 Sep 2011 17:39:56 -0400 Message-ID: <20110916173956.18133qtvqi91m3wg@mail.top-consulting.net> Date: Fri, 16 Sep 2011 17:39:56 -0400 From: freebsd@top-consulting.net To: freebsd-questions@freebsd.org References: <20110916063153.200375qdq59crf8c@mail.top-consulting.net> <32990703-D068-4B0D-AF3A-C1E6EA0A4100@elde.net> <20110916101833.17485ybnq5srjbc4@mail.top-consulting.net> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; format="flowed" Content-Disposition: inline Content-Transfer-Encoding: 7bit User-Agent: Internet Messaging Program (IMP) H3 (4.3.7) X-Originating-IP: 24.200.94.238 X-DSPAM-Result: Innocent X-DSPAM-Processed: Fri Sep 16 17:39:58 2011 X-DSPAM-Confidence: 0.8436 X-DSPAM-Improbability: 1 in 540 chance of being spam X-DSPAM-Probability: 0.0000 X-DSPAM-Signature: 1,4e73c22e11842591782742 X-DSPAM-Factors: 27, wrote+>, 0.00745, ignoring+the, 0.99000, help+>, 0.01000, NFS, 0.01000, NFS, 0.01000, net>+>, 0.01000, the+load, 0.01000, the+load, 0.01000, net+wrote, 0.01000, a+grain, 0.01000, no+point, 0.01000, complain+that, 0.01000, Yes+they, 0.01000, My+point, 0.01000, 1+since, 0.01000, missing+something, 0.01000, in+slower, 0.01000, I've+also, 0.01000, As+per, 0.01900, That's+not, 0.02548, >+On, 0.02929, >+>>, 0.04438, but+according, 0.04537, case+I, 0.04968, of+messages, 0.05116, bad+for, 0.05490, good+fit, 0.05490 X-PolarisMail-Flags: x 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 21:39:59 -0000 Quoting Terje Elde : > 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 then complain that it didn't work. > > Neither can you ask people to donate their time, if you won't spend yours. > > In other words: if you won't listen, there's no point in us talking. > > However: > > Don't disable ZIL. Just don't. It's not the way to go. If you want > to know why, google will help. > > Also, you're making some assumptions, such as the ZIL being bad for > performance. That's not always the case. ZIL-writes are a rather > nice load for spinning metal storage. Even if you write through > cache, that can give you a boost on your real world workload. > > 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 shines at. > > Thanks to the ZIL (the thing you're trying to kill, remember?) you > can convert seek heavy writes to sequential zil-writes, freeing up > disk bandwith for concurrent reads. > > If you want to test before spending money, try what Svein said. Set > up a small logical volume (preferrably smaller than your controller > cache, if it's large enough), then try that as a dedicated zil-device. > > Never tried that, but worth a shot. > > Terje It's not about spending money or not. I really want to use ZFS for some of its features ( journaled, snapshots, etc ) but it has to be a good fit for me. I'm not ignoring the advice I am given, just taking it with a grain of salt disabling the ZIL is recommended - sometimes - for NFS. As per hundreds of messages I've read from the Archive along with this page, http://wiki.freebsd.org/ZFSTuningGuide, it does appear that disabling the ZIL is a solution for NFS. Yes, they still recommend SSD drives and I fully understand that. My point was the following: Why is a sequential write test like dd slower on ZFS than on UFS ? The writes is already serialized so enabling/disabling the ZIL should have very little impact - which is indeed the case. I even went as far as disabling the cache flush option of ZFS through this variable: vfs.zfs.cache_flush_disable: 1, since I already have the write cache of the controller. I've also set some other variables as per the Tuning guide but according to several benchmarks ( iozone, bonnie++, dd ) ZFS still comes in slower than UFS at pretty much everything. Either I am missing something or there is something wrong with my setup.