Re: removing an ext2fs file forces disk activity

[Marcus Brinkmann]

On Tue, Mar 26, 2002 at 11:04:14AM -0700, Jon Arney wrote:
> I understand the goals of having disk syncrhonization performed
> in the proper order to avoid disk inconsistencies.  I also,
> however, agree with Adam that something less than "optimal"
> might be better than nothing at all.

We have something better that is less than optimal.  It is slow but
robust, which is less than optimal (optimal would be fast and robust).
Linux is (probably, haven't checked) fast and not so robust.

Now, if you work on a patch that disables proper synchronization in some of
these cases which are not so critical, as an --unsafe option, we might
consider adding it if the benefit is worth it.

> Perhaps I'll pull out 'bonnie' and some other test programs to
> provide some relative performance numbers under the Hurd v.s.
> Linux just to quantify the trade-off being made.

We are interested in getting hard numbers, but note that comparing Linux
with the Hurd is not a very useful comparison to decide how to improve the
design of the Hurd.  You will have to do some profiling, and maybe try out
the changes you want to make and compare the Hurd without and with the
changes.

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de