[Gluster-devel] IO Scheduling paper - worthy to be considered for GlusterFS ?

[Deepak Shetty]

 Hi,
     I came across this paper ...


http://www.ece.rice.edu/~pjv/mclock.pdf


while going thru a VMWare blog and this is the brief about what it claims
to be :

*In this paper, we presented a novel IO scheduling algorithm,*
*mClock, that provides per-VM quality of service in presence of variable
overall throughput. *
*The QoS re-quirements for a VM are expressed as a minimum reservation, **a
maximum limit, and a proportional share. *

*A key aspect of mClock is its ability to enforce such controls even with
fluctuating overall capacity, as shown by* *our implementation in the
VMware ESX server hypervisor. *

*We also presented dmClock, a distributed version of*
*our algorithm that can be used in clustered storage system architectures. *
*We implemented dmClock in a distributed storage environment and showed *
*that it works as specified, maintaining global per-client reservations,*
*limits, and proportional shares, even though the schedulers run locally *
*on the storage nodes.*

I thought this was worth sharing with the wider audience to see if there is
any value  in looking at dmclock from a GlusterFS perspective ?

thanx,
deepak
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-devel/attachments/20140723/157c7196/attachment-0001.html>