Subject: Re: individual hme sbus cards vs quad hme sbus cards
To: Michael-John Turner <mj@turner.org.za>
From: Bruce O'Neel <edoneel@sdf.lonestar.org>
List: port-sparc
Date: 03/08/2006 12:24:52
Hi,

I have an Ultra II which NFS mounts a disk in a SS20.  The U2
has 2 400mhz UII processors.  The SS20 has 2 180mhz Hypersparcs.
The SS20 has a quad hme and the UII has a single on board hme and talk
to each other through some cheap switch.

I can get approx 4meg/sec off of the SS20 disk when the UII does the
reading.  Going the other direction it slows down to about 1.5meg/sec
writing.

While reading cpu 0 on the ss20 shows about 80% interrupt time.

The same SS20 has two other hme intefaces which I run ipf on.  When
I read from   the UII through the SS20 as the (very basic) firewall
to a Powermac I can get a bit over 2meg/sec.  When the SS20 had 100mhz
processors rather than 180 ones it ran a bit over 1meg/sec.  Once again 
this basically completely occupies cpu 0.

cheers

bruce


On Tue, Mar 07, 2006 at 10:41:10PM +0200, Michael-John Turner wrote:
> On Tue, Mar 07, 2006 at 10:47:55AM +0000, Julian Coleman wrote:
> > What data rate do you expect on your 3 interfaces?
> 
> Hmm... Along those lines - what's the maximum transfer rate anyone has
> achieved with an hme(4) or quad hme(4) in a 32-bit machine? I'd be curious
> to know just how much the faster SuperSPARC-II/HyperSPARC/TurboSPARC-based
> machines can handle, both with and without pf(4)/ipfilter(4).
> 
> -mj
> -- 
> Michael-John Turner | http://weblogs.turner.org.za/mj/
> mj@turner.org.za    | Open Source in WC ZA - http://www.clug.org.za/

-- 
edoneel@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org