> Chuck Silvers <chuq@chuq.com> writes:
> > actually the generational idea isn't a replacement for the knobs.
> > the generational bit still depends on the assumption that past behaviour
> > is a good predictor of future behaviour, so it falls down just like the
> > pre-knob code when that assumption isn't true.
> 
> Absolutely, but *every* policy does. The best part of the research of
> the last 30 years on the subject is the realization that even if you
> have a psychic computer that knows what the entire future of page use
> will be, in the most general case it is still extremely
> computationally expensive to determine what pages to evict.

It is impossible to be perfect - but probably easy to be better!

I was wondering if some sort of 'weighting' based on use count
might let the system discard (some) recent items when it is clear
that the current activitity is requesting lots of new items.

For the 'find' case this might mean that the lower directory nodes
stay resident - as might those the rest of the system has been happily
using.

A similar thing for data pages might stop mkisofs (etc) ditching all
the pages for your desktop.

    David