Am 02.01.2014 18:48, schrieb Dietmar Maurer:
-----Original Message-----
From: Stefan Priebe [mailto:s.priebe@xxxxxxxxxxxx]
Sent: Donnerstag, 02. Jänner 2014 18:36
To: Dietmar Maurer; Dino Yancey
Cc: ceph-users@xxxxxxxxxxxxxx
Subject: Re: rados benchmark question
Hi,
Am 02.01.2014 17:10, schrieb Dietmar Maurer:
Having your journals on the same disk causes all data to be written twice, i.e.
once to the journal and once to the
osd store. Notice that your tested throughput is slightly more than
half your expected maximum...
But AFAIK OSD bench already considers journal writes. The disk can
write about 100MB/s, so 55MB/s is already half speed!
So that does not explain the behavior for me.
Can you please post
iostat -x 5 (for about 30 seconds)
Note: sdb and sdc are my OSDs:
# iostat -x 5 (after about 30 seconds)
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
sdb 0.00 3.80 0.00 187.40 0.00 84663.60 903.56 157.62 796.93 0.00 796.93 5.34 100.00
sdc 0.00 30.00 0.00 169.60 0.00 72501.80 854.97 47.33 279.14 0.00 279.14 5.58 94.62
so your disks are completely utilized and can't keep up see %util and await.
So it's limited by your disks. It would be nice to have the complete 30s
and also a run of rados bench.
May be put it on pastebin or gist.
Stefan
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
