I opened a thread recently here asking about what can be generally
accepted as 'ceph overhead' when using the file system. I wonder if the
performance loss I have on a cephfs 1x replication pool compared to
native performance is really so much. 5,6x to 2x slower than native disk
performance
4k r ran.
4k w ran.
4k r seq.
4k w seq.
1024k r ran.
1024k w ran.
1024k r seq.
1024k w seq.
size
lat
iops
kB/s
lat
iops
kB/s
lat
iops
MB/s
lat
iops
MB/s
lat
iops
MB/s
lat
iops
MB/s
lat
iops
MB/s
lat
iops
MB/s
Cephfs
ssd rep. 3
2.78
1781
7297
1.42
700
2871
0.29
3314
13.6
0.04
889
3.64
4.3
231
243
0.08
132
139
4.23
235
247
6.99
142
150
Cephfs
ssd rep. 1
0.54
1809
7412
0.8
1238
5071
0.29
3325
13.6
0.56
1761
7.21
4.27
233
245
4.34
229
241
4.21
236
248
4.34
229
241
Samsung
MZK7KM480
480GB
0.09
10.2k
41600
0.05
17.9k
73200
0.05
18k
77.6
0.05
18.3k
75.1
2.06
482
506
2.16
460
483
1.98
502
527
2.13
466
489
(4 nodes, CentOS7, luminous)
_____
From: Maged Mokhtar [mailto:mmokh...@petasan.org]
Sent: 15 January 2019 22:55
To: Ketil Froyn; ceph-users@lists.ceph.com
Subject: Re: [ceph-users] Recommendations for sharing a file system to a
heterogeneous client network?
Hi Ketil,
I have not tested the creation/deletion but the read/write performance
was much better then the link you posted. Using CTDB setup based on
Robert's presentation, we were getting 800 MB/s write performance for
queue depth =1 and 2.2 GB/s queue depth= 32 from a single CTDB/Samba
gateway. For the QD=32 test we used 2 Windows clients to the same
gateway (to avoid limitation from the Windows side). Tests were done
using Microsoft diskspd tool at 4M blocks with cache off. Gateway had
2x40 G nics : one for Windows network the other for CephFS client, each
was doing 20 Gbps (50% utilization) cpu was 24 cores running at 85%
utilization taken by the smbd process. We used Ubuntu 16.04 CTDB/Samba
with a SUSE SLE15 kernel for kernel client. Ceph was Luminous 12.2.7.
Maged
On 15/01/2019 22:04, Ketil Froyn wrote:
Robert,
Thanks, this is really interesting. Do you also have any details on
how a solution like this performs? I've been reading a thread about
samba/cephfs performance, and the stats aren't great - especially
when creating/deleting many files - but being a rookie, I'm not 100%
clear on the hardware differences being benchmarked in the mentioned
test.
http://lists.ceph.com/pipermail/ceph-users-ceph.com/2018-May/026841.h
tml
Regards, Ketil
On Tue, Jan 15, 2019, 16:38 Robert Sander
<r.san...@heinlein-support.de wrote:
Hi Ketil,
use Samba/CIFS with multiple gateway machines clustered with CTDB.
CephFS can be mounted with Posix ACL support.
Slides from my last Ceph day talk are available here:
https://www.slideshare.net/Inktank_Ceph/ceph-day-berlin-unlimited-
fileserver-with-samba-ctdb-and-cephfs
Regards
--
Robert Sander
Heinlein Support GmbH
Schwedter Str. 8/9b, 10119 Berlin
https://www.heinlein-support.de
Tel: 030 / 405051-43
Fax: 030 / 405051-19
Amtsgericht Berlin-Charlottenburg - HRB 93818 B
Geschäftsführer: Peer Heinlein - Sitz: Berlin
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
--
Maged Mokhtar
CEO PetaSAN
4 Emad El Deen Kamel
Cairo 11371, Egypt
www.petasan.org
+201006979931
skype: maged.mokhtar
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
