From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753480Ab0IWGXO (ORCPT ); Thu, 23 Sep 2010 02:23:14 -0400 Received: from dion.org.ua ([193.104.35.153]:45525 "EHLO dion.org.ua" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1750731Ab0IWGXN convert rfc822-to-8bit (ORCPT ); Thu, 23 Sep 2010 02:23:13 -0400 Date: Thu, 23 Sep 2010 06:32:59 +0000 From: Dmitry Nezhevenko To: Ryan Castellucci Cc: linux-kernel@vger.kernel.org Subject: Re: dm-crypt and huge performance penalty Message-ID: <20100923063259.GA18250@dion.org.ua> References: <20100920223700.GA20507@laptop.local> MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline Content-Transfer-Encoding: 8BIT In-Reply-To: User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Sep 22, 2010 at 04:52:17PM -0700, Ryan Castellucci wrote: > On Mon, Sep 20, 2010 at 3:37 PM, Dmitry Nezhevenko wrote: > > Firstly I've tried to setup raid5 and then dm-crypt on top of it. After > > discovering issue I've tried just raid5 and confirmed that mdadm itself > > works as expected. > > > > And the last idea is to setup three dm-crypt partitions and then raid5 on > > top of independently encrypted drives. > > > > So here are my results: > > > > š š š š š | šmdadm š | mdadm+dm-crypt | 3 x dm-crypt + mdadm | > > Seq. read š| 168 Mb/s | 57 Mb/s š š š š| 119 Mb/s š š š š š š | > > Seq. write | 80 Mb/s š| 36 Mb/s š š š š| 64.4 Mb/s š š š š š š| > > dm-crypt will only use one core per device, so this is expected > behavior (and I believe an issue that is being worked on). From these > numbers it looks like you have a dual core or dual cpu system? > Hi, Yes, The machine is home NAS with inexpensive Intel Core i3 CPU that has two real cores and HT. -- WBR, Dmitry