[Gluster-users] GlusterFS Performance gigE
douglas.m.stanley at gmail.com
Fri Sep 10 17:28:29 UTC 2010
Wow, that doesn't sound right! I have a gluster cluster at home using
old P4 hardware, and cheap soho off the shelf gigabit switches, and I
can get ~35MB/s writes with dd:
dd if=/dev/zero of=/vol/scratch/foo/test.img bs=1M count=64
64+0 records in
64+0 records out
67108864 bytes (67 MB) copied, 1.76719 s, 38.0 MB/s
dd if=/dev/zero of=/vol/scratch/foo/test.img bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 29.3242 s, 36.6 MB/s
And that's with no tuning of gigabit, i.e. no jumbo frames or
anything, everything is stock, and off the shelf. One of my machines
might even be using an old pci 33mhz bus cheap realtek gigabit card.
The above numbers were on a straight distribute between two nodes, 1
brick each. On my other volume that's doing replicate between the two
nodes, I get about 18MB/s (same hardware as above).
On Fri, Sep 10, 2010 at 12:16 PM, Henrique Haas <henrique at dz6web.com> wrote:
> Hello people,
> For those have *GlusterFS running over GigE, what is the throughput you have
> for write data?*
> I have 3 nodes, running GNU/Linux Ubuntu Server 10.04, GlusterFS 3.0.2.
> The network is GigE, dedicated switch.
> I am experiencing maximum 2 MB/sec running a "dd" over a client mount point.
> Thank you very much
> Henrique Haas
> +55 (51) 3028.6602
> Gluster-users mailing list
> Gluster-users at gluster.org
Please avoid sending me Word or PowerPoint attachments.
More information about the Gluster-users