On Tue, Jul 12, 2016 at 8:24 PM, Gandalf Corvotempesta <gandalf.corvotempesta@xxxxxxxxx> wrote:
Generally most of these systems have a dedicated metadata server where for each file they store on which servers(replicas) the file is stored. In gluster there is no metadata server, while it gives flexibility in some other ways, In this aspect it becomes a bit rigid. The steps I gave above eventually achieve the same thing. Another way to solve this problem is to flip the order of replication and distribution layers but that will need a lot of subsystems to be tested thoroughly.2016-07-12 14:06 GMT+02:00 Anuradha Talur <atalur@xxxxxxxxxx>:
> I realize I just repeated the same example in different words but I hope this answers
> your questions. So do let us know if something is not clear.
Is unclear, probably because i don't understand how gluster works and
because I don't
understand why was designed like this. Having to replace bricks just
to add a single
server and not multiple servers due to the replica count, is absolutely unclear.
Any other scale-out storage allows this natively.
--
Pranith
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users
