Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 26 Apr 1997 15:55:28 -0700
From:      David Greenman <dg@root.com>
To:        dennis <dennis@etinc.com>
Cc:        "Michael K. Sanders" <msanders@aros.net>, Christopher Sedore <cmsedore@mailbox.syr.edu>, freebsd-hackers@FreeBSD.ORG, isp@FreeBSD.ORG
Subject:   Re: Router statistics 
Message-ID:  <199704262255.PAA25437@root.com>
In-Reply-To: Your message of "Sat, 26 Apr 1997 12:30:10 EDT." <3.0.32.19970426123001.00b16380@etinc.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
>Now I assume that this machine has a rather heavy overhead with all of the
>users
>and disk activity....do you have any estimates of the impact of this on the
>overall
>networking throughput? Obviously the SCSI activity is going to suck up much
>bus bandwidth.....

   The numbers are packet sends/receives to user processes throught the
TCP/IP stack. I would expect the packet forwording capability to be much
greater. The majority of the CPU time is spent doing file related things,
not networking.

>As for the Intel Pro/100B ...is this a 10/100MB device? Does it have separate
>TP connectors, or 1? Are there any clones that are supported, or any versions
>of it that are not supported?

   10/100, one connection, one vendor. The one vendor is a good thing,
however, because the design isn't "pot luck" and the device driver is
much less complicated as well. Or in other words, good for users, bad
for Intel haters. The chip is available from Intel for people if they
want to make clone cards, however.

-DG

David Greenman
Core-team/Principal Architect, The FreeBSD Project




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704262255.PAA25437>