From owner-freebsd-questions@freebsd.org Tue May 16 10:34:49 2017 Return-Path: Delivered-To: freebsd-questions@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id BC11CD6F102 for ; Tue, 16 May 2017 10:34:49 +0000 (UTC) (envelope-from jim@ohlste.in) Received: from maurice.jlkmail.com (maurice.jlkmail.com [23.111.151.218]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 71A081EB0 for ; Tue, 16 May 2017 10:34:49 +0000 (UTC) (envelope-from jim@ohlste.in) Received: from maurice.jlkmail.com (maurice.jlkmail.com [127.0.0.1]) by maurice.jlkmail.com (Postfix) with ESMTP id E757E33AAEE7 for ; Tue, 16 May 2017 06:34:41 -0400 (EDT) Authentication-Results: maurice.jlkmail.com (amavisd-new); dkim=pass (1024-bit key) reason="pass (just generated, assumed good)" header.d=ohlste.in DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ohlste.in; h= content-transfer-encoding:content-language:content-type :content-type:in-reply-to:mime-version:user-agent:date:date :message-id:from:from:references:to:subject:subject; s=dkim; t= 1494930879; x=1495794880; bh=lfOiWOZk8mblf5WaKw75+KodwjlYioT1STt Vc0IgMEE=; b=qshUpio6LCwfLousai0dRPJBlteQlZ5BHOd0mhePqLMtHu45kZj GqC+QqWokNe4QRvIWGZiIFDSPkvWNOIz0vKL06XiTkIqzmIxrCOsIDETCCeMfufJ bcRcpflXY8cUSGZRs4CBQiZpzdFvnKXkX9rTIJ0ZcyeFefjwLLDwdBFQ= X-Virus-Scanned: Debian amavisd-new at maurice.jlkmail.com X-Spam-Flag: NO X-Spam-Score: -1 X-Spam-Level: X-Spam-Status: No, score=-1 tagged_above=-100 required=6.31 tests=[ALL_TRUSTED=-1] autolearn=unavailable autolearn_force=no Received: from maurice.jlkmail.com ([127.0.0.1]) by maurice.jlkmail.com (maurice.jlkmail.com [127.0.0.1]) (amavisd-new, port 10026) with ESMTP id zxbGlc-Fy-hd for ; Tue, 16 May 2017 06:34:39 -0400 (EDT) Received: from [192.168.1.153] (static-70-104-198-154.nrflva.fios.verizon.net [70.104.198.154]) by maurice.jlkmail.com (Postfix) with ESMTPSA id 47B4033AAB45; Tue, 16 May 2017 06:34:39 -0400 (EDT) Subject: Re: ZFS root on single SSD? To: Aaron References: <99fa2537-9fb1-0ccf-d906-39db1c2e2685@FreeBSD.org> From: "jim@ohlste.in" Cc: freebsd-questions@freebsd.org, Matthew Seaman Message-ID: <9997fd01-273c-b176-b9ed-e33e9e2d1b2f@ohlste.in> Date: Tue, 16 May 2017 06:34:38 -0400 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.0 MIME-Version: 1.0 In-Reply-To: <99fa2537-9fb1-0ccf-d906-39db1c2e2685@FreeBSD.org> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Language: en-US Content-Transfer-Encoding: 8bit X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 May 2017 10:34:49 -0000 Hello, On 05/16/2017 03:39 AM, Matthew Seaman wrote: > On 16/05/2017 06:45, Aaron wrote: >> So, I've been running ZFS root mirror across 2 spinning disks, and I'm >> upgrading my home server/nas and planning on running root on a spare SSD. >> However, I'm unsure if it'd be better to run UFS as a single drive root >> instead of ZFS, although I do love all of the ZFS features (snapshots, COW, >> scrubbing, etc) and would still like to keep that for my root drive, even >> if I'm not mirroring at all. I do notice that FreeBSD has TRIM support for >> ZFS (see http://open-zfs.org/wiki/Features#TRIM_Support). >> >> So is there a good reason NOT to run ZFS root on a single drive SSD? > > No. Running ZFS on a single device works fine, although you obviously > don't benefit from all the really nice resilience features. > > The choice between UFS2 and ZFS basically comes down to three points: > > * performance -- for certain IO patterns, UFS can out-perform ZFS > quite markedly. Particularly the sort of small, randomly distributed > IOs you get with a RDBMS. Of course, for database use, the additional > data security you get from ZFS makes it desirable despite this. > > * system resources -- ZFS is memory hungry. This is not a problem on > most contemporary machines, which tend to have sufficient RAM, but older > machines, VMs or appliances may struggle. > > * data security -- the integrated checksumming in ZFS provides > assurance that the data you're reading now is the same as what you wrote > previously. Now, this is almost always the case with UFS2 (would be > entirely useless if not), but there is no actual guarantee of it, and > silent data corruption is possible[*]. If you're handling data which is > really important or in particularly large volumes or where your hardware > may prove deficient, then ZFS is indicated. > > Cheers, > > Matthew > > [*] With only one drive and one copy of each file, ZFS cannot provide > resilience against data errors, but it will prevent it going unnoticed. > I'd add only that while a mirrored zpool offers some data protection, it is *not* an effective "backup" solution for important data. Drive failure during resilver after a drive replacement does occur. If there's important data on the drive, backing it up to a different medium is still essential, whether it's a mirrored pool or a single drive pool. [Leaned once the hard way]™ -- Jim Ohlstein