On Tue, Dec 16, 2003 at 08:47:08AM +0900, YAMAMOTO Takashi wrote:
> hi,
> 
> > On Tue, Dec 09, 2003 at 03:21:18PM +0100, Manuel Bouyer wrote:
> > > I did a test with solaris and NetBSD:
> > > the test was
> > > time dd if=/dev/zero of=toto bs=128k count=10000
> > > time dd if=toto bs=128k of=/dev/null
> > > I ran these 5 times, and did an average of the time returned, excluding
> > > times clearly too high (the server is in production, and the 100Mb/s
> > > link is shared with other systems).
> > > The server is a NetBSD/alpha 1.6.1_STABLE, with an array of 6 SCSI disks on 2
> > > SCSI busses for the partition I ran the tests on, and a gigabit network
> > > interface. It has 20 nfsds running.
> > > 
> > > The solaris client is running solaris9 on a UltraSPARC-IIi 333MHz with 256MB
> > > RAM. Default mount options (remote/read/write/nosuid/hard/intr/xattr).
> > > The NetBSD client is running current as of today, on a UltraSPARC-IIi
> > > 400Mhz with 256MB RAM. Also using amd's default mount options
> > > (nfs (nosuid, nodev)).
> 
> - what's your network environment?  higher latency than normal 100baseTX link?

It's 100baseTX but is has to go though several swicthes to the server (nothing
unusual I guess, for large networks)

> - "default" means, netbsd uses udp and solaris uses tcp?

Solaris is probably using TCP. I'm using amd, so I don't know what NetBSD is
using. 

> - how about cpu usage?

Comparable (but note that the solaris box has a slower CPU). CPU doens't
semm to be the bottle neck.

> - do you have packet dumps or such?

No, and I'm away from paris for 2 weeks, so I can't get some right now.

> 
> 
> FYI, the following is a result on my notebook.
> (p3-700, 384M memory, 100baseTX, TCP, 4 nfsiods)
> 
> kaeru% time dd if=/dev/zero of=a bs=128k count=8000
> 8000+0 records in
> 8000+0 records out
> 1048576000 bytes transferred in 94.801 secs (11060811 bytes/sec)
> dd if=/dev/zero of=a bs=128k count=8000  0.17s user 15.25s system 16% cpu 1:34.81 total
> kaeru% time dd of=/dev/zero if=a bs=128k count=8000
> 8000+0 records in
> 8000+0 records out
> 1048576000 bytes transferred in 99.565 secs (10531572 bytes/sec)
> dd of=/dev/zero if=a bs=128k count=8000  0.10s user 10.23s system 10% cpu 1:39.78 total

What hardware do you have on the server side ?


-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 23 ans d'experience feront toujours la difference
--