Florin Iucha wrote:
On Thu, Mar 10, 2005 at 09:21:30AM -0500, Shailabh Nagar wrote:


+The I/O controller consists of +- a new I/O scheduler called ps-iosched which is an incremental update +to the cfq ioscheduler. It has enough differences with cfq to warrant a
+separate I/O scheduler. +- ckrm-io : the controller which interfaces ps-iosched with CKRM's core


I do not like the entangling of the specific scheduler (ps) data into
the ckrm io controller since that will prevent the reuse of that module
and lead to duplicate code and confused administrators.

As I am writing a io scheduler based on cello (I am mostly learning and
making sense of it all), I would like to use the same ckrm controller.
To that end, what would be needed is the ckrm io controller to hold
bald pointers to scheduler specific structures, much like the elevator
and the request have private data members that can then hold the
queues, semaphores and other fields.

What do you think?

Yes, that could be done. It creates another interface level (b/w the I/O controller and the I/O scheduler) and now requires us to consider whether we permit multiple CKRM-compliant/aware schedulers to coexist. I think it can be done...just have to make sure that it works.


-- Shailabh



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech

Reply via email to