On Wed, Dec 03, 2003 at 01:43:30PM +1030, Greg 'groggy' Lehey wrote:
> On Thursday, 27 November 2003 at 18:02:52 +0100, Olaf Seibert wrote:
> > However, when I test with a benchmark called RawIO (see
> > http://www.acnc.com/benchmarks.html) on the (unused) raw swap partition,
> > I get much better results. Sequential writing is around 35M/sec, and the
> > number of xfers/sec is over 1000, peaking at some 1900.
> 
> This is surprisingly high.  You can assume that you're not getting
> valid results.

Uh, that's nonsense.  I have several PCI RAID controllers that can
deliver at least that level of performance; one of them is almost four
years old.  The card in cvs.netbsd.org, which is identical to the one
Olaf has, performs about as well as he's reporting -- the difference,
though, is that we're not seeing the performance degradation through
the filesystem that he is.  The reason for that is something I haven't
managed to figure out, unfortunately.

Thor