Hello fellow Ceph users,
We have been using a small cluster (6 data nodes with 12 disks each, 3
monitors) with OSDs on spinners and journals on SATA SSD-s for a while now.
We still haven't upgraded to Luminous, and are going to test it now, as
we also need to switch some projects on a shared file system and cephFS
seems to fit the bill.
What I'm mostly looking for is to get in contact with someone with
experience in running Ceph as a full SSD cluster, or full SSD pool(s) on
the main cluster. Main interest is in performance centric workloads
generated by web applications that work directly with files, heavily
both in read and write capacity, with low latency being very important.
As mentioned above, the other question is about viability of cephFS in
production environment right now, for web applications with several
nodes, using a shared file system for certain read and write operations.
I will not go into more detail here, if you have some experience and
would be willing to share it, please write to val...@eenet.ee
Also thanks to everyone in this list for the insights other people's
random problems have given us. We have probably managed to prevent some
problems in the current cluster just by skimming through these e-mails.
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
