Re: Ceph RDMA setting for public/cluster network

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

 



On 15:55 Thu 26 Sep, Roman Penyaev wrote:
> > I'll write patch to remove "ms_async_rdma_device_name" and get the
> > device name through public_addr/cluster_addr.
> 
> Removal is not a good option since you always have to think about
> compatibility.
I agree with that compatibility is quite important. However, if the
feature isn't widely used in production environment, we could make some
change to improve it.
> 
> > BTW, have you seen user use deploy Ceph/RDMA in production environment?
> > Do they seperate the public/cluster network or use the same network?
> 
> Never met the separation network for rdma transport.  But I can be
> wrong here.
Do you mean that you've seen that some users use Ceph/RDMA in production
environment with using same network for public/cluster message? If the
RDMA NIC has 2 ports, how to make full use of the ports in this kind of
case?
> 
> --
> Roman
> 
_______________________________________________
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