From owner-freebsd-questions@FreeBSD.ORG Mon Apr 11 10:58:49 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 D01951065672 for ; Mon, 11 Apr 2011 10:58:49 +0000 (UTC) (envelope-from freebsd-questions@m.gmane.org) Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by mx1.freebsd.org (Postfix) with ESMTP id 8A9578FC08 for ; Mon, 11 Apr 2011 10:58:49 +0000 (UTC) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1Q9EpU-0007cQ-JX for freebsd-questions@freebsd.org; Mon, 11 Apr 2011 12:58:48 +0200 Received: from lara.cc.fer.hr ([161.53.72.113]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Mon, 11 Apr 2011 12:58:48 +0200 Received: from ivoras by lara.cc.fer.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Mon, 11 Apr 2011 12:58:48 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-questions@freebsd.org From: Ivan Voras Date: Mon, 11 Apr 2011 12:58:33 +0200 Lines: 19 Message-ID: References: <4DA131BD.3030805@telting.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: lara.cc.fer.hr User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101102 Thunderbird/3.1.6 In-Reply-To: <4DA131BD.3030805@telting.org> X-Enigmail-Version: 1.1.2 Subject: Re: ZFS Striping and Optimizing Capabilities 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: Mon, 11 Apr 2011 10:58:49 -0000 On 10/04/2011 06:27, Chris Telting wrote: > Just a few questions about what ZFS actually does. So if anyone has > intimate knowledge about ZFS's implementation on Freebsd I'm sure I and > others would appreciate the answers. > > When you add a second and or thrid drive/partition to a zpool I'm > assuming that it's going to start using the drives like a raid 0 stripe. Not exactly like RAID 0; ZFS's "striping" is not using regular patterns / stripes but "IO balancing" heuristics. This means there are no fixed stripe size but the data is distributed across the drives in a way which ZFS's algorithms predict is best. This allows for e.g. creating a zpool from a single drive, then adding another drive (as a "stripe") and ZFS will use both, but new data will be written mostly to the new drive. Performance should scale as expected in this scenario, e.g. not much if the first drive is almost full before adding a new one.