Return-Path: Resent-Message-ID: <20170227150220.vpmbghkbxaufyumm@eorzea.usersys.redhat.com> Resent-To: sandeen@sandeen.net Received: from mr001msb.fastweb.it ([85.18.95.85]:38504 "EHLO mr001msb.fastweb.it" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751115AbdBXL7t (ORCPT ); Fri, 24 Feb 2017 06:59:49 -0500 Received: from ceres.assyoma.it (93.63.55.57) by mr001msb.fastweb.it (8.5.140.05) id 5890BA1F010F5161 for linux-xfs@vger.kernel.org; Fri, 24 Feb 2017 12:47:29 +0100 From: Gionatan Danti Subject: Increase sync write performance? Message-ID: Date: Fri, 24 Feb 2017 12:47:23 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: linux-xfs@vger.kernel.org Cc: g.danti@assyoma.it Hi list, I am trying to understand if, and how, to increase synchronous write performance to an XFS filesystem. A common scenario is an NFS-exported directory with sync writes enabled (default). With EXT4, it is possible to mount it using the "data=journal" option which, by eliminating small seeks, give a very nice performance boost (at least with short write spikes). AFAIK, XFS only journals metadata, never user data alone. There is nothing I can do to increase its sync write performance in this case? Short of using a powerloss-protected controller/drives with writeback caches enabled, of course... Thanks. -- Danti Gionatan Supporto Tecnico Assyoma S.r.l. - www.assyoma.it email: g.danti@assyoma.it - info@assyoma.it GPG public key ID: FF5F32A8