From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alasdair G Kergon Subject: Re: [linux-lvm] 2 LVM questions - LVM and 2.4.19, and extent based striping. Message-ID: <20021126222800.E6509@uk.sistina.com> References: <20021126213244.D6509@uk.sistina.com> Mime-Version: 1.0 Content-Disposition: inline In-Reply-To: <20021126213244.D6509@uk.sistina.com>; from agk@uk.sistina.com on Tue, Nov 26, 2002 at 09:32:44PM +0000 Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Tue Nov 26 16:28:02 2002 List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-lvm@sistina.com On Tue, Nov 26, 2002 at 09:32:44PM +0000, Alasdair Kergon wrote: > If you have 18 extents (PEs) with 6 on each disk numbered: > disk1: 0 1 2 3 4 5 > disk2: 6 7 8 9 10 11 > disk3: 12 13 14 15 16 17 [More on the numbering LVM* uses, in case that wasn't clear enough.] Depending whereabouts those extents are on the disk, the *actual* PE numbers that LVM* assigns could be completely different (but still consecutive on each disk inside each segment of the LV) e.g. disk1: 0 1 2 3 4 5 disk2: 50 51 52 53 54 55 disk3: 8 9 10 11 12 13 And the corresponding logical extent (LE) numbers within the logical volume run *along* each stripe of the segment in turn: disk1: 0 1 2 3 4 5 disk2: 6 7 8 9 10 11 disk3: 12 13 14 15 16 17 So in LE terms, the data get striped across 0, 6 and 12, then 1, 7 and 13 etc., equivalent to striping across 3 extents each 6 times bigger, just one per disk viz. [0..5], [6..11] and [12..17]. Alasdair -- agk@uk.sistina.com