From: Carlos Maiolino Subject: [LSF/MM ATTEND] filesystem - SMR drives Date: Mon, 13 Jan 2014 13:06:44 -0200 Message-ID: <20140113150643.GC22358@orion.maiolino.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org To: lsf-pc@lists.linux-foundation.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:1244 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751619AbaAMPHK (ORCPT ); Mon, 13 Jan 2014 10:07:10 -0500 Content-Disposition: inline Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hi, giving the new SMR technology, and the current performance penalty we might see with some write patterns (mainly random-write related), I'd like to join into a discussion about what kind of filesystem improvements we can implement to reduce the re-write performance impact on these disks and, of course, help implementing them. I'm mostly interested in ext4 and xfs since these are the filesystems I actively work on, but I'm willing to help/contribute with another filesystems that also need changes on their write patterns. -- Carlos