Possible performance improvement?
A message that is not in the archive (sent off-list or not yet received): [email protected]
---------- Forwarded message ----------
From: Pranith Kumar <[email protected]>
Date: Tue, Feb 9, 2010 at 2:33 PM
Subject: Possible performance improvement?
To: Ingo Molnar <[email protected]>, LKML <[email protected]>
Hey Ingo,
I recently happened to come across this thesis[1] which presents some very concrete results with stats regarding the performance of Linux CFS with an implementation of Redline.
It would really be great if you could take some time out of your busy schedule and quickly look up the figures to see the performance comparison and if we could adapt some ideas. I am willing to give it a try if you think its worth it.
Thanks,
--
Pranith.
[1] http://scholarworks.umass.edu/cgi/viewcontent.cgi?article=1071&context=open_access_dissertations
Quoted 10 lines · Pranith Kumar
I recently happened to come across this thesis[1] which presents some very concrete results with stats regarding the performance of Linux CFS with an implementation of Redline.
It would really be great if you could take some time out of your busy schedule and quickly look up the figures to see the performance comparison and if we could adapt some ideas. I am willing to give it a try if you think its worth it.
[1] http://scholarworks.umass.edu/cgi/viewcontent.cgi?article=1071&context=open_access_dissertations
Looks to propose a rather simple semi-static edf based cpu scheduler.
If you're interested in that, there is a group actively working on a deadline scheduler for linux: http://lkml.org/lkml/2009/10/16/161