CEPH filesystem development
 help / color / mirror / Atom feed
From: Michael Metz-Martini | SpeedPartner GmbH <metz@speedpartner.de>
To: Christian Balzer <chibi@gol.com>, ceph-users@ceph.com
Cc: Sage Weil <sage@redhat.com>,
	ceph-devel@vger.kernel.org, ceph-maintainers@ceph.com
Subject: Re: [ceph-users] Deprecating ext4 support
Date: Wed, 13 Apr 2016 14:51:58 +0200	[thread overview]
Message-ID: <570E40EE.40202@speedpartner.de> (raw)
In-Reply-To: <20160413112930.49906f15@batzmaru.gol.ad.jp>

Hi,

Am 13.04.2016 um 04:29 schrieb Christian Balzer:
> On Tue, 12 Apr 2016 09:00:19 +0200 Michael Metz-Martini | SpeedPartner
> GmbH wrote:
>> Am 11.04.2016 um 23:39 schrieb Sage Weil:
>>> ext4 has never been recommended, but we did test it.  After Jewel is
>>> out, we would like explicitly recommend *against* ext4 and stop
>>> testing it.
>> Hmmm. We're currently migrating away from xfs as we had some strange
>> performance-issues which were resolved / got better by switching to
>> ext4. We think this is related to our high number of objects (4358
>> Mobjects according to ceph -s).
> It would be interesting to see on how this maps out to the OSDs/PGs.
> I'd guess loads and loads of subdirectories per PG, which is probably where
> Ext4 performs better than XFS.
A simple ls -l takes "ages" on XFS while ext4 lists a directory
immediately. According to our findings regarding XFS this seems to be
"normal" behavior.

pool name       category                 KB      objects
data            -                       3240   2265521646
document_root   -                     577364        10150
images          -                96197462245   2256616709
metadata        -                    1150105     35903724
queue           -                  542967346       173865
raw             -                36875247450     13095410

total of 4736 pgs, 6 pools, 124 TB data, 4359 Mobjects

What would you like to see?
tree? du per Directory?

As you can see we have one data-object in pool "data" per file saved
somewhere else. I'm not sure what's this related to, but maybe this is a
must by cephfs.

-- 
Kind regards
 Michael Metz-Martini

  parent reply	other threads:[~2016-04-13 12:52 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-11 21:39 Deprecating ext4 support Sage Weil
     [not found] ` <alpine.DEB.2.11.1604111632520.13448-Wo5lQnKln9t9PHm/lf2LFUEOCMrvLtNR@public.gmane.org>
2016-04-11 21:42   ` Allen Samuels
2016-04-11 21:47     ` [ceph-users] " Jan Schermer
2016-04-11 23:39   ` Christian Balzer
2016-04-12  1:12     ` [ceph-users] " Sage Weil
     [not found]       ` <alpine.DEB.2.11.1604112046570.29593-Wo5lQnKln9t9PHm/lf2LFUEOCMrvLtNR@public.gmane.org>
2016-04-12  1:32         ` Shinobu Kinjo
2016-04-12  2:05         ` [Ceph-maintainers] " hp cre
2016-04-12  2:43       ` [ceph-users] " Christian Balzer
2016-04-12 13:56         ` Sage Weil
     [not found]           ` <alpine.DEB.2.11.1604120837120.29593-Wo5lQnKln9t9PHm/lf2LFUEOCMrvLtNR@public.gmane.org>
2016-04-13  3:27             ` Christian Balzer
     [not found]     ` <20160412083925.5106311d-9yhXNL7Kh0lSCLKNlHTxZM8NsWr+9BEh@public.gmane.org>
2016-04-14  9:43       ` Antw: " Steffen Weißgerber
2016-04-12  7:00   ` Michael Metz-Martini | SpeedPartner GmbH
2016-04-13  2:29     ` [ceph-users] " Christian Balzer
2016-04-13 12:30       ` Sage Weil
2016-04-14  0:57         ` Christian Balzer
2016-04-13 12:51       ` Michael Metz-Martini | SpeedPartner GmbH [this message]
2016-04-11 21:44 ` Sage Weil
2016-04-11 21:57   ` Mark Nelson
     [not found]     ` <570C1DBC.3040408-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-04-11 22:49       ` Shinobu Kinjo
2016-04-11 23:54         ` [ceph-users] " Robin H. Johnson
2016-04-11 23:09       ` Lionel Bouton
2016-04-12  7:45   ` [ceph-users] " Jan Schermer
2016-04-12 18:00     ` Sage Weil
2016-04-12 19:19       ` Jan Schermer
2016-04-12 19:58         ` Sage Weil
2016-04-12 20:33           ` Jan Schermer
2016-04-12 20:47             ` Sage Weil
     [not found]               ` <alpine.DEB.2.11.1604121639590.29593-Wo5lQnKln9t9PHm/lf2LFUEOCMrvLtNR@public.gmane.org>
2016-04-12 21:08                 ` Nick Fisk
     [not found]                   ` <4f0f087c.9Ro.9Gf.hg.1qX1VyMOaD-ImYt9qTNe79BDgjK7y7TUQ@public.gmane.org>
2016-04-12 21:22                     ` wido-fspyXLx8qC4
     [not found]               ` <362740c3.9Ro.9Gf.hg.1qX1VyMOaE@mailjet.com>
2016-04-12 23:12                 ` [ceph-users] " Jan Schermer
2016-04-13 13:13                   ` Sage Weil
2016-04-13 13:06               ` Sage Weil
2016-04-14 18:05                 ` Jianjian Huo
2016-04-14 18:30                   ` Samuel Just
2016-04-12  6:39 ` [Ceph-maintainers] " Loic Dachary
2016-04-13 14:19 ` [ceph-users] " Francois Lafont

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=570E40EE.40202@speedpartner.de \
    --to=metz@speedpartner.de \
    --cc=ceph-devel@vger.kernel.org \
    --cc=ceph-maintainers@ceph.com \
    --cc=ceph-users@ceph.com \
    --cc=chibi@gol.com \
    --cc=sage@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox