Re: Disk priorities...

[Rik van Riel]

On Sun, 1 Oct 2000, Alexander Viro wrote:
> On Sun, 1 Oct 2000, Rik van Riel wrote:
> > On Sun, 1 Oct 2000, LA Walsh wrote:
> > 
> > > Forgive me if this has been asked before, but has there ever
> > > been any thought of having a 'nice' value for disk accesses?.  
> > 
> > Not currently, but it would be trivial to adjust the maximum
> > elevator sorting latency according to the niceness of the
> > process. I have no idea how much this would help, though ...
> 
> What process? By the time when bh hits the ll_rw_block()
> originating process may be dead.

Indeed, for write throttling we have to find another
solution. One thing we could do is set different
trottle targets for different processes.

(so process A is trottled at 10% dirty pages while
process B is trottled at 40% dirty pages ... with
priorities being recalculated once every second or
so to preserve fairness)

> And if you mean reads... Good luck propagating the originator
> information.

Isn't it the case that for most of the filesystem
reads the current process is the one that is the
originator of the request ?

regards,

Rik
--
"What you're running that piece of shit Gnome?!?!"
       -- Miguel de Icaza, UKUUG 2000

http://www.conectiva.com/		http://www.surriel.com/