[Gluster-devel] afr, striping, and nufa scheduler

Krishna Srinivas krishna at zresearch.com
Tue Sep 25 17:25:47 UTC 2007


Hi August,

Actually the read scheduler is not yet done, it is on our roadmap to
implement it. Now, AFR does all the read operations from its first
child.

Regards
Krishna

On 9/25/07, August R. Wohlt <glusterfs at isidore.net> wrote:
> Hi,
>
> I have a setup with two servers mirroring each others' files using
> cluster/afr:2. The two servers are on a high-latency connection, so I would
> like to ensure that the local server is given preference in i/o over the
> distant server. I do not need aggregrate performance, only file replication.
>
> Is it the default behavior to do i/o on the first server listed in the afr
> configration before the 2nd? I see this in the FAQ about striping, which
> makes me wonder:
>
>   GlusterFS AFR (automatic file replication) translator does a striped read
> to improve performance on mirrored files.
>
> And I do not see the equivalent of a "scheduler" option in cluster/afr in
> the src that would let me use something like NUFA.
>
> What is the best way to ensure this?
>
> Thanks!
> :august
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>





More information about the Gluster-devel mailing list