We were able to saturate hardware with EC as well. Could you check 'top' in threaded mode to see if fuse thread is saturated when you run dd?
On Mon, Apr 24, 2017 at 3:27 PM, Ingard Mevåg <ingard@xxxxxxxx> wrote:
HiI've been playing with disperse volumes the past week, and so far i can not get more than 12MB/s when i do a write test. I've tried a distributed volume on the same bricks and gotten close to gigabit speeds. iperf confirms gigabit speeds to all three servers in the storage pool.The three storage servers have 10gig nics (connected to the same switch). The client is for a now a docker container in a 2nd DC (latency roughly 8-9 ms).dpkg -l|grep -i glusterii glusterfs-client 3.10.1-ubuntu1~xenial1 amd64 clustered file-system (client package)ii glusterfs-common 3.10.1-ubuntu1~xenial1 amd64 GlusterFS common libraries and translator modulesii glusterfs-server 3.10.1-ubuntu1~xenial1 amd64 clustered file-system (server package)$ gluster volume infoVolume Name: DFS-ARCHIVE-001Type: DisperseVolume ID: 1497bc85-cb47-4123-8f91-a07f55c11dcc Status: StartedSnapshot Count: 0Number of Bricks: 1 x (4 + 2) = 6Transport-type: tcpBricks:Brick1: dna-001:/mnt/data01/brickBrick2: dna-001:/mnt/data02/brickBrick3: dna-002:/mnt/data01/brickBrick4: dna-002:/mnt/data02/brickBrick5: dna-003:/mnt/data01/brickBrick6: dna-003:/mnt/data02/brickOptions Reconfigured:transport.address-family: inetnfs.disable: onAnyone know the reason for the slow speeds on disperse vs distribute?kind regardsingard
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
--
Pranith
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users