cluster detail：
ceph version 0.94
3 host, 3 mon, 18 osd
1 ssd as journal + 6 hdd per host.
1 pool, name is rbd , pg_num is 1024, 3 replicated.
step：
1.
rbd create test1 -s 81920
rbd create test2 -s 81920
rbd create test3 -s 81920
2.
on host1, rbd map test1, get /dev/rbd0 on kernel 3.18.11 or /dev/rbd1 on kernel 3.13.6
on host2, rbd map test2, get /dev/rbd0 on kernel 3.18.11 or /dev/rbd1 on kernel 3.13.6
on host3, rbd map test3, get /dev/rbd0 on kernel 3.18.11 or /dev/rbd1 on kernel 3.13.6
3.
startup IOMeter1.1.0, 24 workers per host, total 72 workers, 1 outstanding, 5s ramp up time, 120s run time
order: 4KB seq read, 4KB seq wirte, 4KB rand read, 4KB rand write
4.getting these numbers in results.csv
Thank you!
yangruifeng
-----邮件原件-----
发件人: Ilya Dryomov [mailto:idryomov@xxxxxxxxx]
发送时间: 2015年4月14日 15:44
收件人: yangruifeng 09209 (RD)
抄送: ceph-users
主题: Re: rbd performance problem on kernel 3.13.6 and 3.18.11
On Tue, Apr 14, 2015 at 6:24 AM, yangruifeng.09209@xxxxxxx <yangruifeng.09209@xxxxxxx> wrote:
> Hi all！
>
>
>
> I am testing rbd performance based on kernel rbd dirver, when I
> compared the result of the kernel 3.13.6 with 3.18.11, my head gets so confused.
>
>
>
> look at the result, down by a third.
>
>
>
> 3.13.6 IOPS
>
> 3.18.11 IOPS
>
> 4KB seq read
>
> 97169
>
> 23714
>
> 4KB seq write
>
> 10110
>
> 3177
>
> 4KB rand read
>
> 7589
>
> 4565
>
> 4KB rand write
>
> 10497
>
> 2307
How exactly are you getting these numbers? Which tools are you using, which tests are you running, in which order?
Thanks,
Ilya
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
