[Gluster-devel] AFR scheduling
krishna at zresearch.com
Wed Jun 18 19:41:44 UTC 2008
Ah! that kind of read scheduling already happens now. i.e a file is read from
a particular subvol depending on the inode number, this gives
us a fair scheduling. This is by default. However if you want to
read from a particular subvol for all files you can specify it by
"option read-subvolume <volname>"
On Thu, Jun 19, 2008 at 12:46 AM, <jcanter at clemson.edu> wrote:
> The thread I am referring to is http://lists.gnu.org/archive/html/gluster-devel/2007-10/msg00085.html
> I am specifically referring to the section
> In 1.4 we will have:
> * feature to allow scheduling of read() operations amongst the
> subvols in round-robin fashion.
> We wish to be able to do some load balancing been our different volumes. Right now all of our
> clients read from the same server, which causes a bit of a backup. Thanks.
>> There are two meanings for read scheduling:
>> 1) already implemented, load balance reads such that a file is always read
>> from the same subvol.
>> 2) not implemented, load balance a read such that, for a read call of 1000 bytes
>> 500 bytes is read from 1st subvol and next 500 bytes is read from
>> 2nd subvol,
>> they are aggregated and given to the application.
>> Which read scheduling did you mention about?
>> Do you have the mail thread which talked about it?
>> On Wed, Jun 18, 2008 at 11:38 PM, <jcanter at clemson.edu> wrote:
>>> I've been poking around with gluster for some research work and found a post about a future
>>> implementation of afr read scheduling for version 1.4 release. This is the specific feature
>>> we are looking for based on our cluster setup and research needs. Is there any timeframe for
>>> release of this feature or a way for it to become available? Thank you.
>>> Joshua Canter
>>> Gluster-devel mailing list
>>> Gluster-devel at nongnu.org
More information about the Gluster-devel