Re: Clients' connection for concurrent access to ceph

[John Spray <john.spray@xxxxxxxxxx>]

On 22/07/15 20:39, Shneur Zalman Mattern wrote:

 

Third test:

    We wanted to try CephFS, because our client is familiar with Lustre, that's very near to CephFS capabilities:

           1. I've used my CEPH nodes in the client's role. I've mounted CephFS on one of nodes, and ran dd with bs=1M ...

                    - I've got wonderful write performance ~ 1.1 GBytes/s (really near to 10Gbit network throughput)

            

            2. I've connected CentOS client to 10gig public network, mounted CephFS, but ...

                    - It was just ~ 250 MBytes/s

            

            3. I've connected Ubuntu client (non-ceph member) to 10gig public network, mounted CephFS, and ...

                    - It was also ~ 260 MBytes/s

            Now I have to know: perhaps ceph-members-nodes have privileged access ???

While you're benchmarking, it's a good idea to try both the kernel client and the fuse client.  You may find one works better than the other, and we'll find the numbers interesting too.

You're using giant, the latest LTS release is hammer -- if you're interested in cephfs you'll be better off with hammer (lots of new stuff going in all the time).

Aside from that, it's kind of surprising that your servers are working as better clients than your clients.  Do the clients definitely have all the same kernel+ceph versions as the servers?  Is the link between the clients and the servers definitely 10G all the way across your network?  Is a pure network benchmark seeing the full 10gbps between a client node and a server node?

Cheers,
John

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com