[ceph-users] Cluster network slower than public network

Gandalf Corvotempesta gandalf.corvotempesta at gmail.com
Wed Nov 15 04:58:34 PST 2017

Small info: all of our nodes (3 to 5 plus 6 hypervisors) have 4 10gb ports
but we only have 2 10gb switches (small port numbers, only 16, so we can't
place both network on the same switch)

We use 2 switches for HA in active-backup mode

I was thinking to use both 10gb switches as public network (the one
connected to the hypervisors) and two 1gb switch as cluster network

Il 15 nov 2017 1:50 PM, "Gandalf Corvotempesta" <
gandalf.corvotempesta at gmail.com> ha scritto:

> As 10gb switches are expansive, what would happen by using a gigabit
> cluster network and a 10gb public network?
> Replication and rebalance should be slow, but what about public I/O ?
> When a client wants to write to a file, it does over the public network
> and the ceph automatically replicate it over the cluster network or the
> whole IO is made over the public?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20171115/520c6c9a/attachment.html>

More information about the ceph-users mailing list