From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx.bordeaux.inra.fr ([147.100.111.220]:38771 "EHLO mail-1.bordeaux.inra.fr" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753701AbbFPG1y (ORCPT ); Tue, 16 Jun 2015 02:27:54 -0400 Received: from [147.100.98.136] ([147.100.98.136]) (authenticated bits=0) by mail-1.bordeaux.inra.fr (/8.14.4) with ESMTP id t5G6Rp6s007342 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 16 Jun 2015 08:27:51 +0200 (CEST) Message-ID: <557FC1E7.6040108@bordeaux.inra.fr> Date: Tue, 16 Jun 2015 08:27:51 +0200 From: Tovo Rabemanantsoa MIME-Version: 1.0 To: Btrfs BTRFS Subject: Re: Bad performance with near-full FS References: <557EC95C.8020902@bordeaux.inra.fr> <20150615132915.GM6761@twin.jikos.cz> <557EE4BF.2090908@bordeaux.inra.fr> <557EF12E.6000309@gmail.com> In-Reply-To: <557EF12E.6000309@gmail.com> Content-Type: text/plain; charset=utf-8 Sender: linux-btrfs-owner@vger.kernel.org List-ID: On 06/15/2015 05:37 PM, Austin S Hemmelgarn wrote: > For almost any non-COW filesystem (ext4, XFS, JFS, etc.), 1% or 100MB > (whichever is larger) is generally a good buffer. On BTRFS, I would say > at least 5% or 1.5G (again, whichever is larger; and if performance is a > concern, go for at least 10-20%), as BTRFS is known to have some rather > poor behavior when running very close to full. > > I got it. Thanks all ! T.