Frank van der Linden wrote:

> On Wed, Jan 24, 2001 at 09:22:06AM -0500, Lennart Augustsson wrote:
> > I'm running with your patch on my laptop now and the difference
> > is very noticable.  Now I can actually do something else while
> > doing a `cvs update' or compiling.
>
> That's cool, but: I'd like to see some kind of a benchmark that
> gives comparable numbers. I mean: maybe this change will make
> behavior the same as with the traditional buffer cache again :-)

Absolutely!  I wasn't suggesting that my subjective feeling of speed
was going to determine if UBC's performance is adequate.  :-)
I was just expressing my joy over being able to use my machine again.

I'm not sure how to make benchmarks for all of these things.  How do
you test interactive responsiveness in a benchmark (without having
someone clicking and clocking)?  I'm sure it's possible, but it will
require some thought.

    -- Lennart