Subject: Re: UVM/other problems for desktop users in current?
To: NetBSD-current Discussion List <current-users@netbsd.org>
From: Dan Melomedman <dan%dan.dan@devonit.com>
List: current-users
Date: 12/18/2002 11:19:56
Peter Seebach wrote:
> In message <20021218070721.GB21205@pyy.jmp.fi>, Jukka Marin writes:
> >Next they'll blame it on IDE, sure, but I don't buy that, either - I've
> >got only one disk per IDE port and the disks are using Ultra/100 DMA.
> >The disks are fast (but RAID appears to kill performance).
> 
> >I'm also seeing the problems in interactive use on client systems, I'm
> >just saying you don't need X to notice these problems.
> 
> Okay, let's turn this around:
> 
> Anyone seeing this on SCSI?  Anyone able to reproduce it on SCSI?
> 
> If we're seeing horrible performance botches on one-drive-per-port IDE
> systems (and I am, and you are), but we can't find them on SCSI, we have
> a culprit.
> 
> Despite all the griping about this, I think we're onto something positive
> that could improve the system substantially.

I don't have NetBSD on it, but...

I have a FreeBSD 1 GHz Athlon machine with 256 MB of RAM, Mylex
controller, and two 10K RPM 18G SCSI drives in a mirror. Building world
or updating the source tree using cvsup doesn't affect interactive
performance much. Mozilla and XMMS are humming along fine. Of course
it's even faster when 16 MB of write-back cache is on. I can finally
forget most of the time about waiting for the drives to finish
their thing. No kernel tuning, softupdates are on.