I would be surprised if a P133 has any significant bottleneck, at
least if you keep it relatively lightly loaded. Suppose you actually
can get 1MB/s: That is about 700 ethernet packets/second (an ethernet
packet is 1500 bytes.) A PII/300 appears able to handle about 10000
ethernet packets/second (sending and receiving each packet.) A P133
is probably only about 5 times slower. You would be more likely to
see a bottleneck on a 100Mb/s network, but as far as I can tell DNO
doesn't know how to configure a 100Mb/s network anyway. And you
probably cannot get 1MB/s if you are going any significant distance
(e.g., to the research side of DNO's network.)
-Bradley
Date: Mon, 25 Oct 1999 08:26:14 -0400 (EDT)
From: Michael Osier <mosier@krunch.med.yale.edu>
Reply-To: Michael Osier <michael.osier@yale.edu>
On Sun, 24 Oct 1999, Bradley C. Kuszmaul wrote:
> It may depend on just how old and slow the linux box is.
>
> I never had any troubles with using a PII 300 for a router on a
> 10-base-T network.
That's the problem. My linux box is a P 133. Roughly where would you say
the cutoff would be?
> If performance is the only issue, I would be surprised if it is a big
> enough issue to matter.
I need to do a lot of large file transfers (5 Mb+ files) for part of my
research. The transfers are a bit of a bottleneck to my work. Thus my
concern. If I had to make a choice, I'd wide up having to dump my linux
box in favor of the Win box. :(
I'm also looking to pick up a 2nd Win box so I can work on one while the
other is processing. If I do, I'd need two boxes with fast connections.
Thus the questions about speed with IP Masq. What would be the minimal
system configuration you would recommend for IP Masq. for a fast
connection?
Michael
----------------------------------------------------------------------
Michael Osier | "He is not well rounded who does
http://krunch.med.yale.edu/~og/ | not have an equally keen interest
BS Biochemical Science - UVM | in all of the things within the
Yale University | compass of painting."
Human Genetics - Og | Leonardo da Vinci
nospamog@krunch.med.yale.edu
This archive was generated by hypermail 2b29 : Wed Apr 27 2005 - 03:30:03 EDT