Date: Mon, 03 May 1999 19:19:30 -0600 From: Wes Peters <wes@softweyr.com> To: dg@root.com Cc: sthaug@nethelp.no, hackers@FreeBSD.ORG Subject: Re: Decent network cards for 100Mbit? Message-ID: <372E4B22.4EC5B758@softweyr.com> References: <199905032318.QAA04559@implode.root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
David Greenman wrote:
>
> Wes Peters wrote:
> >
> >I should probably point out I'm doing network throughput torture
> >tests with 64-byte packets. ;^) Any reasonably good Fast Ethernet
^^^^^^^^^^^^^^^^^^^^
> You benchmark is broken. The pro/100, for one, can easily do 100Mbps
> continuously. I should know since not only did I write the driver, but I
The benchmark isn't broken, it's just weird. No card can do 100Mbps with
64-byte packets, the preamble lops off 1/3 of your performance using those
itty bitty packets. The PNIC came closer to the theoretical max than anything
else.
> use it exclusively here and on wcarchive (and in the latter, I have graphs
> and download stats to prove it).
IIRC, the EEPro has significantly lower interrupt overhead at 100 Mbps
than the Tulip, or probably the PNIC. For a web or ftp server, that's
a big plus. For a multicast router, the EEPro is probably going to
suffer somewhat vs. cards that can hash all valid ethernet multicast
addresses. At least that's the assumption we're running on. One of
the systems has an EEPro on-board in it, when we get the full test
running I'll try it on both interfaces and see how the EEPro does
against the PNIC. I assume the 3c905, with a single-bit control for
multicast, is going to suck. The 3c905B seems to have varying degrees
of support for multicast filtering, according to comments in the driver.
--
"Where am I, and what am I doing in this handbasket?"
Wes Peters Softweyr LLC
http://www.softweyr.com/~softweyr wes@softweyr.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?372E4B22.4EC5B758>