--- On Thu, 5/27/10, Metin Akyalı <metin.akyali@xxxxxxxxx> wrote:
> I would use 1gbit client servers and 100mbit storage
> servers. All the servers will be in same DC. I will
> rent the servers and i dont own my own DC house. Reason
> i am choosing 1gbit server as the client is i
> wont have too much 1gbit server, but i will have many
> stogage nodes, and 1gbit server is very expensive but
> 100mbit is not so.
It sounds like you may not really need replication
if all the files have to pass through the "client".
Simply add a caching layer to the client, I suspect
that most FSes can do that. I know that glusterfs
certainly can, they have a caching translator.
But, even plain old NFS should likely cache enough
on the client side if it is a read only load.
If not, AFS does provide some read only specific
replication, but I think that you have to specify
which files to replicate. From what I understand
the ceph clients will also cache data locally with a
read only load.
Additionally, both ceph and glusterfs (at least) will
stripe files across multiple storage nodes, so even
without replication they should be able to serve up
a single file at a higher bandwidth than the
connection to one single node. I am not sure that
replication really is the solution you seek.
My $.02,
-Martin
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
