Matthias Scheler writes:
> In article <41EE207F.3000708@netbsd.ro>,
> 	Mihai CHELARU <kefren@netbsd.ro> writes:
> > Shortly, RaidFrame is performing very poor on NetBSD 2.0.
> 
> Because you asked for poor performance. ;-)
> 
> > Here is my configuration:
> [...]
> > START layout
> > 32 1 1 5
> 
> RAID 5 in hardware is slow. RAID 5 in software is very, very slow.

Yes.. very, very slow.. I only hit 138MB/sec reads and 105MB/sec 
writes to a RAIDframe RAID 5 set...  Truly pathetic ;)

     -------Sequential Output-------- ---Sequential Input-- --Random--
     -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
  MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
1500 86487 76.4 108263 70.5 10128  8.0 90153 74.4 141263 44.4 356.7  4.3

(NetBSD 2.0E-ish 2.8GHz Xeon i386, RAIDframe, 5 U320 Seagate SCSI 
drives on an Adaptec 29160 controller.  64K/8K block/frags FFSv1)

> This is not specific to RAIDframe but to every RAID 5 software
> implementation.

Obviously, I don't agree with your generalization. :)

(Of course, I'm completely ignoring application-level stuff here, 
and the benchmark might not be perfect, but since we're just 
hand-waving, I figured I can play too ;) )

Later...

Greg Oster