From mboxrd@z Thu Jan 1 00:00:00 1970 From: Adam Goryachev Subject: Re: RAID performance Date: Thu, 07 Feb 2013 21:41:06 +1100 Message-ID: <511384C2.7020601@websitemanagers.com.au> References: <51134E43.7090508@websitemanagers.com.au> <5113516C.5080606@aei.mpg.de> <51137DFA.5020406@websitemanagers.com.au> <51138206.5000301@aei.mpg.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <51138206.5000301@aei.mpg.de> Sender: linux-raid-owner@vger.kernel.org To: Carsten Aulbert Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On 07/02/13 21:29, Carsten Aulbert wrote: > Hi > > On 02/07/2013 11:12 AM, Adam Goryachev wrote: >> Do you have any information on what your workload is, or how/why these >> values might help? > > Most of our data servers are exporting data read-only via NFS. It helped > well on these but more on download servers which allowed us to go from > 200k 4MB downloads per day to 1M/day while adding new files constantly. > >> >> You are changing values significantly from the default, and I am >> cautious that they may cause other issues. Also, someone else has >> advised to reduce nr_requests rather than increasing it? >> > > I know, but these larger queues really helped a lot in re-ordering > requests to better match the hardware underneath - but again, this was > for hard drives with physical arms and not SSDs OK, this makes a lot more sense then :) Definitely slowing things down/making the opportunity to re-order requests should help a lot for spinning disks... I don't see that helping much at all with SSD's though. Thanks, Adam -- Adam Goryachev Website Managers www.websitemanagers.com.au