Re: wild imbalance in kswapd state

Matthias Urlichs (smurf@noris.de)
18 Aug 1998 08:19:26 +0200


"Stephen C. Tweedie" <sct@redhat.com> writes:
> On Mon, 20 Jul 1998 16:04:52 -0400, Bill Hawes <whawes@star.net> said:
> > The results were pretty interesting -- once the swapping starts, it
> > stays parked in the shrink_mmap state until almost all of the page cache
> > is gone.
>
> Yep. All the benchmarking on machines <=16MB shows that this is
> required behaviour or performance sinks like a rock.
>
On the other hand, with early 2.0 kernels and stock (i.e., ancient ;-) INN,
our Netnews server went twice as fast when I forced a state transition
after every allocation.

The fact that it really didn't have enough memory for both INN and squid
might have something to do with that.

The effect is gone now, of course. (squid is gone; INN with cyclic storage
to a small number of huge files; and the dcache -- the problem would
probably have gone away with each of these changes alone. ;-)

> It may be reasonable to allow the user to tune the behaviour to allow
> more caching on larger memory machines, but when memory gets low, being
> aggressive on the caches definitely appears to be the right thing to
> do.
>

-- 
Matthias Urlichs
noris network GmbH

- 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.altern.org/andrebalsa/doc/lkml-faq.html