[ceph-users] Cluster network slower than public network

Gandalf Corvotempesta gandalf.corvotempesta at gmail.com
Wed Nov 15 05:19:56 PST 2017


Any idea?
I have 1 16 ports 10gb switch, 2 or more 24ports gigabit switches and 5
OSDs (MONs running over them) and 5 hypervisor servers to connect to the
storage

At least 10 ports are needed for each network, thus, 20 ports for both
cluster and public, right ?

I don't have 20 10gb ports....

Il 15 nov 2017 2:10 PM, "Micha Krause" <micha at krausam.de> ha scritto:

> Hi,
>
> Replication and rebalance should be slow, but what about public I/O ?
>>
>
> Writes to clients are only acked after replication, so you can't take
> advantage of your fast public network.
>
> http://docs.ceph.com/docs/giant/_images/ditaa-54719cc959473e
> 68a317f6578f9a2f0f3a8345ee.png
>
> Micha Krause
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20171115/8f4397b8/attachment.html>


More information about the ceph-users mailing list