From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Priebe Subject: Re: btrfs big metadata Date: Mon, 02 Jul 2012 19:40:09 +0200 Message-ID: <4FF1DCF9.6050604@profihost.ag> References: <4FF1461D.9060304@profihost.ag> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.profihost.ag ([85.158.179.208]:42026 "EHLO mail.profihost.ag" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752687Ab2GBRkO (ORCPT ); Mon, 2 Jul 2012 13:40:14 -0400 In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Gregory Farnum Cc: "ceph-devel@vger.kernel.org" Am 02.07.2012 18:42, schrieb Gregory Farnum: > On Sun, Jul 1, 2012 at 11:56 PM, Stefan Priebe - Profihost AG > wrote: >> Hello list, >> >> i found several people who use big metadata -n 64k -l 64k for ceph. But i >> haven't found any ceph doc or info why to use this? >> >> What's the reason to use big metadata feature with ceph? > > One of the big problems users have had with Ceph-on-btrfs is that > Ceph's usage patterns cause btrfs to dramatically fragment its > metadata. Big metadata blocks reduce this fragmentation to a level > where some users find it faster than xfs (which it previously wasn't). Thanks! Stefan