Re: FreeBSD vs. Linux (another question)

["Stephen J. Turnbull" <turnbull@example.com>]

>>>>> "Fredric" == Fredric Fredricson <fredric.fredriksson@example.com> writes:

    Fredric> I have been told by some people (that I think should
    Fredric> know) that FreeBSD disk I/O is faster than Linux. I don't
    Fredric> know but this sound like superstition to me.

Of course it's not superstition.  There are any number of parameters
you can tweak in a caching, read-ahead, etc, etc I/O system.  If
FreeBSD is better than Linux at virtual memory management (not hard,
from what I hear), the reduction in page thrash alone would speed up
other disk I/O perceptibly.

As for adapting the free code, of course they can and do.  Certainly
for hardware drivers like ATAPI and SCSI.  But the interaction with
the virtual file system can surely slow things down, and that is not
easy to synch the code at all.  (IIRC, it's not until 2.4 that your
average kernel driver is allowed to do DMA to a user-space buffer,
which means extra copying.  If such double and triple buffering is a
"feature" of the Linux VFS, that would slow it down too.)

    Fredric> Is there any truth to the claim?

That I can't speak to.  The theory's obvious, though.


-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
_________________  _________________  _________________  _________________
What are those straight lines for?  "XEmacs rules."