Re: Re: [RFC][PATCH 5/9] CPU controller - Documents how the controller works

Re: Re: [RFC][PATCH 5/9] CPU controller - Documents how the controller works

[Martin Andersson]

 > A better way to achieve the desired group cpu usage IMHO would be to
 > adjust nice level of members at slice refresh time.  This way, you get
 > the timeslice scaling and priority adjustment all in one.
 >
 > (I think I would do both actually, with nice level being preferred such
 > that dynamic priority spread within the group isn't flattened, which can
 > cause terminal starvation within the group, unless really required.)

I am actually working on something more or less like that as part of my 
licentiate (half time PhD) thesis. In the thesis I want to control CPU 
resources using a control theoretic approach. The results so far are 
very promising, but my code is not very mature yet. I plan to have my 
thesis ready during this summer, so if anyone is interested I can mail 
you a copy. If I have the time after that, I will try to incorporate my 
code into the CKRM framework.

/Martin Andersson