> From: osymh@gemini.oscs.montana.edu (Michael L. Hitch)
> 
> On Apr 24,  9:29pm, "Stephen J. Roznowski" wrote:
> > > From: osymh@gemini.oscs.montana.edu (Michael L. Hitch)
> ...
> > >  Also, the equivalent of what Sysinfo does would probably be:
> > > 	dd if=/dev/rsd0c of=/dev/null bs=1024k count=10
> > 
> > Just tried that -- got ~2.5Mbytes/sec now. I'm running with an 8K/1K
> > filesystem. Is a 16k one that much faster?
> 
> [ . . . ]
> 
>   I also suspect that tweaking the file system parameters probably
> isn't going to make a whole lot of difference in the overall performance.
> A lot of the I/O is likely to be small transfers (512, 1K, 4K, or 8K),
> and if there is more than one process doing I/O, the I/O will be scattered
> over the disk.  I see much more improvement in I/O performance by using
> a higher performance disk.  NetBSD runs much better on my Maxtor MXT-540SL
> than it does on my Maxtor 7345SY.

Well, it's a little disappointing to be giving up 4-6x the raw performance
of my drive....

A few more tidbits: maxconfig is currently set to 1; I'll try the following
and let people know the results:

	1. measure performance
	2. change rotdelay to 0ms from 4ms
	3. measure
	4. change maxconfig from 1 to 8
	5. measure

I'm doing most of this testing on a Quantum LP52S drive that has a 4.4BSD
filesystem on it. After I'm sure nothing is going to go wrong, I'll try
this on a Quantum PD1050iS drive that has 4.2/4.3BSD filesystems on it.

Thanks for all the info,
-SR