Re: [2.1.130-3] Page cache DEFINATELY too persistant... feature?

Rogier Wolff (R.E.Wolff@BitWizard.nl)
Sat, 28 Nov 1998 14:44:06 +0100 (MET)

Linus Torvalds wrote:
> > This looks like a natural way to take advantage of existing behaviour to
> > fix a very long standing write performance problem. Comments? Would
> > you prefer a more natural way of throttling writes at source?
>
> I think throwttling them at the source is just wrong, and think your
> suggestion makes sense. I don't know whether it makes much sense to do
> this with bdflush, though - I'd be more inclined to just do it directly.
>
> Think of it as the analogous case to kswapd - we count on kswapd to
> generally keep "enough" memory free and to balance the peak usage. But
> when we really need more memory for a process, the process will try to
> free stuff up itself.

I'm not sure that this is possible. We will need to "throttle at the
source" eventually. I can write an application that writes a file at
50Mb per second (try "dd if=/dev/zero of=testfile"), and the disk is
never going to keep up with that.

So working on the "flush to disk" side is good and very important, we
will never be able to ignore the "throttle the writes".

Or am I being stupid?

Roger. 

-- 
My pet light bulb is a year old today.   \_________  R.E.Wolff@BitWizard.nl
That's 5.9*10^12 miles. Your mileage will NOT vary.\__Phone: +31-15-2137555
--(time <-> distance can be converted: lightspeed)--  \____ fax: ..-2138217
We write Linux device drivers for any device you may have! \_______________


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/