Re: cephfs, low performances

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 20 December 2015 at 15:06, Francois Lafont <flafdivers@xxxxxxx> wrote:
Hi,

On 20/12/2015 19:47, Don Waterloo wrote:

> I did a bit more work on this.
>
> On cephfs-fuse, I get ~700 iops.
> On cephfs kernel, I get ~120 iops.
> These were both on 4.3 kernel
>
> So i backed up to 3.16 kernel on the client. And observed the same results.
>
> So ~20K iops w/ rbd, ~120iops w/ cephfs.

Ok. Please, can you give us your configuration?
How many nodes, osds, ceph version, disks (SSD or not, HBA/controller), RAM, CPU, network (1Gb/10Gb) etc.?

All nodes have 10Gbps to each other
OSD:
$ ceph osd tree
ID WEIGHT  TYPE NAME        UP/DOWN REWEIGHT PRIMARY-AFFINITY 
-1 5.48996 root default                                       
-2 0.89999     host nubo-1                                    
 0 0.89999         osd.0         up  1.00000          1.00000 
-3 0.89999     host nubo-2                                    
 1 0.89999         osd.1         up  1.00000          1.00000 
-4 0.89999     host nubo-3                                    
 2 0.89999         osd.2         up  1.00000          1.00000 
-5 0.92999     host nubo-19                                   
 3 0.92999         osd.3         up  1.00000          1.00000 
-6 0.92999     host nubo-20                                   
 4 0.92999         osd.4         up  1.00000          1.00000 
-7 0.92999     host nubo-21                                   
 5 0.92999         osd.5         up  1.00000          1.00000 

Each contains 1 x Samsung 850 Pro 1TB SSD (on sata)

Each are Ubuntu 15.10 running 4.3.0-040300-generic kernel.
Each are running ceph 0.94.5-0ubuntu0.15.10.1  

nubo-1/nubo-2/nubo-3 are 2x X5650 @ 2.67GHz w/ 96GB ram.
nubo-19/nubo-20/nubo-21 are 2x E5-2699 v3 @ 2.30GHz, w/ 576GB ram.

the connections are to the chipset sata in each case.
The fio test to the underlying xfs disk
(e.g. cd /var/lib/ceph/osd/ceph-1; fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=readwrite --filename=rw.data --bs=4k --iodepth=64 --size=5000MB --readwrite=randrw --rwmixread=50)
shows ~22K IOPS on each disk.

nubo-1/2/3 are also the mon and the mds:
$ ceph status
    cluster b23abffc-71c4-4464-9449-3f2c9fbe1ded
     health HEALTH_OK
            election epoch 1104, quorum 0,1,2 nubo-1,nubo-2,nubo-3
     mdsmap e621: 1/1/1 up {0=nubo-3=up:active}, 2 up:standby
     osdmap e2459: 6 osds: 6 up, 6 in
      pgmap v127331: 840 pgs, 6 pools, 144 GB data, 107 kobjects
            289 GB used, 5332 GB / 5622 GB avail
                 840 active+clean
  client io 0 B/s rd, 183 kB/s wr, 54 op/s


_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux