[Gluster-devel] Gluster resource scheduling

[Raghavendra G]

Hi,
afr supports an option "read-subvolume". Reads are by default scheduled to
the node set in this option. Hence In your configuration, each client can
use a different value for the above option to spread out the reads to
different servers.

regards,
On Wed, Jun 25, 2008 at 9:31 PM, <jcanter at clemson.edu> wrote:

> Hi,
>
> I have a cluster with 10 servers setup in afr mode(set from client side)
> which may be called upon
> by any of 14 clients.  When I attempt to read from the machines, they all
> try to pull data from a
> single server.  Is there a way to configure the load to be spread out among
> the mirrored servers
> rather than on a single one?  Would server side afr work better in such a
> case?  My setup is as
> follows:
>
> #Client Setup
> volume server1
>  type protocol/client
>  option transport-type tcp/client
>  option remote-host server1
>  option remote-subvolume brick
> end-volume
> ...
> volume server10
>  type protocol/client
>  option transport-type tcp/client
>  option remote-host server10
>  option remote-subvolume brick
> end-volume
>
> volume mirror0
>  type cluster/afr
>  subvolumes server1 server2 server3 server4 server5 server6 server7 server8
> server9 server10
> end-volume
>
>
>
> #server config
> volume brick
>  type storage/posix
>  option directory /dfs/
> end-volume
>
> volume server
>  type protocol/server
>  option transport-type tcp/server
>  option auth.ip.brick.allow *
>  subvolumes brick
> end-volume
>
>
> Thank you
>
> Josh
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>



-- 
Raghavendra G

A centipede was happy quite, until a toad in fun,
Said, "Prey, which leg comes after which?",
This raised his doubts to such a pitch,
He fell flat into the ditch,
Not knowing how to run.
-Anonymous