What is your Ceph version? From the test results you posted, your environment's
performance is okay in regard of your setup. But there are definitely many
things that can be tuned to get you better number.
I normally use top, iostat, pidstat, vmstat, dstat, iperf3, blktrace, netmon,
ceph admin
Can you post the fio results with the ioengine using libaio? From what you
posted, it seems to me that the read test hit cache. And the write performance
was not good, the latency was too high (~35.4ms) while the numjobs and iodepth
both were 1. Did you monitor system stat on both side (VM/Compu
Hi Jim, when you do reweighting, balancing will be triggered, how did you set
it back? Setting back immediately or waiting for balancing to complete? I did
try both on my cluster and couldn't see osd bench changed significantly like
yours (actually no changes), however, my cluster is 12.2.12, no
Hi Jim, did you check system stat (e.g. iostat, top, etc.) on both osds when
you ran osd bench? Those might be able to give you some clues. Moreover, did
you compare both osds' configurations?
-- Original --
From:
aha, thanks very much for pointing out, Anthony!
Just a summary for the screenshot pasted in my previous email. Based on my
understanding, "ceph daemon osd.x ops" or "ceph daemon osd.x
dump_ops_in_flight" shows the ops currently being processed in the osd.x. I
also noticed that there is anothe
"ceph daemon osd.x ops" shows ops currently in flight, the number is different
from "ceph osd status
we have a cluster with very low load, however, "ceph osd perf" shows high
commit_latency and apply_latency.
root@stor-mgt01:~# ceph -s
cluster:
id:
3d1ec789-829d-4e0f-b707-9363356a68f1
health: HEALTH_WARN
application
not enabled on 3 pool(s)
services:
mon: 3 d
72360.338257,
"iops": 2773.626104
}
root@stor-mgt01:~# ceph tell osd.30 bench 98304000 32768
{
"bytes_written": 98304000,
"blocksize": 32768,
"elapsed_sec": 2.908703,
"bytes_per_sec": 33796507.598640,
"iops": 1031.387561
}
root@stor-mgt01:~# ceph tell osd.30 bench 49152000 16384
{
"bytes_written": 49152000,
"blocksize": 16384,
"elapsed_sec": 3.907744,
"bytes_per_sec": 12578102.861185,
"iops": 767.706473
}
-- --
??:
"rainning"
Hi Zhenshi,
I did try with bigger block size. Interestingly, the one whose 4KB osd bench
was lower performed slightly better in 4MB osd bench.
Let me try some other bigger block sizes, e.g. 16K, 64K, 128K, 1M etc, to see
if there is any pattern.
Moreover, I did compare two SSDs, they respec
one more thing it seems that WAL does have more impact on small write.
---Original---
From: "Zhenshi Zhou"
Hi Zhenshi, thanks very much for the reply.
Yes I know it is ood that the bluestore is deployed only with a separate db
device but no a WAL device. The cluster was deployed in k8s using rook. I
was told it was because the rook we used didn't support that.
Moreover, the comparison was made on
Hi all,
I am wondering if there is any performance comparison done on osd bench with
and without a separate WAL device deployed given that there is always a
separate db device deployed on SSD in both cases.
The reason I am asking this question is that we have two clusters and osds in
one hav
12 matches
Mail list logo
