On Mon, Mar 2, 2015 at 7:15 PM, Nathan O'Sullivan <nat...@mammoth.com.au> wrote:
>
> On 11/02/2015 1:46 PM, 杨万元 wrote:
>
> Hello!
>     We use Ceph+Openstack in our private cloud. Recently we upgrade our
> centos6.5 based cluster from Ceph Emperor to Ceph Firefly.
>     At first,we use redhat yum repo epel to upgrade, this Ceph's version is
> 0.80.5. First upgrade monitor,then osd,last client. when we complete this
> upgrade, we boot a VM on the cluster,then use fio to test the io
> performance. The io performance is as better as before. Everything is ok!
>     Then we upgrade the cluster from 0.80.5 to 0.80.8,when we  completed ,
> we reboot the VM to load the newest librbd. after that we also use fio to
> test the io performance.then we find the randwrite and write is as good as
> before.but the randread and read is become worse, randwrite's iops from
> 4000-5000 to 300-400 ,and the latency is worse. the write's bw from 400MB/s
> to 115MB/s. then I downgrade the ceph client version from 0.80.8 to 0.80.5,
> then the reslut become  normal.
>      So I think maybe something cause about librbd.  I compare the 0.80.8
> release notes with 0.80.5
> (http://ceph.com/docs/master/release-notes/#v0-80-8-firefly ), I just find
> this change in  0.80.8 is something about read request  :  librbd: cap
> memory utilization for read requests (Jason Dillaman)  .  Who can  explain
> this?
>
>
> FWIW we are seeing the same thing when switching librbd from 0.80.7 to
> 0.80.8 - there is a massive performance regression in random reads.   In our
> case, from ~10,000 4k read iops down to less than 1,000.
>
> We also tested librbd 0.87.1 , and found it does not have this problem - it
> appears to be isolated to 0.80.8 only.

I'm not familiar with the details of the issue, but we're putting out
0.80.9 as soon as we can and should resolve this. There was an
incomplete backport or something that is causing the slowness.
-Greg
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to