In message <311F144DB5E5D011B03F00805FE954B10687835D@exchange01_ph.border=
s.com>
          "Calvin Vette (IT- Borders Online)" <CVETTE@borders.com> wrote:=


> Once I decoded your names :-) -=20
>=20
> You've got NetBSD running on the slowest machines, with ISA Ethernet ca=
rds.
> This is going to make a huge difference. And while you didn't mention i=
t,
> the amount of RAM could make a difference in low memory conditions. My =
point
> is, it's not a very scientific comparison.
>=20
> I'd recommend testing a little more scientifically to get a real compar=
ison
> - try swapping the NetBSD 1.3.2 drives into the NT box (Proc & RAM).=20
>=20
> If it's just raw networking speed you're after, try using NetBSD curren=
t
> with UVM. (A release since the beginning of September). Chuck's pretty =
much
> eliminated excessive internal copying, which goes a long way towards
> improving Network I/O (and in UVM's case, all I/O).
Sorry, but I think this won't help Olaf much.

I've played a little bit with samba.1.9.18p10 (=3Dlatest at the time of
writing) an NetBSD-current. I just can't get more than ~1/3 of the
transfer rates I get with ftp (~340kB/s vs 1.04MB/s) on a switched
10BaseT/100BaseTX network.

I 've tried to tune the samba server and played with various compile
flags and configuration options mentioned in docs/Speed.txt in the samba
source-tree (I didn't use the samba-package) - didn't changed much
(nearly nothing).

I've used xosview to monitor the cpu/net/disc activity:
It looks like the bandwith usage goes down after ~1/2 second to the
level I've mentioned above, CPU and Disc usage are low whole time.

Well, I suppose Samba & NetBSD are just no big friends (at the
moment...)

Bye,
 Thilo.
--=20
Mir ist mein Signature entlaufen :-(. Wer es findet, sende es bitte an
Thilo.Manske@HEH.Uni-Oldenburg.DE zur=FCck. Danke!=20