Internet Backbone Index

Joe Shaw jshaw at
Fri Jun 27 12:44:38 UTC 1997

I would think that as long as he's got some UN*X variant/flavor on there,
that a p90 would be enough.  Now, had he put Microsoft NT/IIS on there, it
probably wouldn't be able to handle the os, much less running web
services.  You'd be suprised at how silly a UN*X flavor can make an NT box
look when you compare them side by side.

Hell, I know someone that runs 4 heavily traveled non-profit websites on a
486sx33 with 4 megs of ram and a couple of megs of hardrive space running 
FreeBSD. They get a couple a decent amount of hits a day, and never even
burp.  When people find out that it's only a 486sx33, they are usually
quite impressed.  All I can say is that you could not do that with a
microsoft product.  

Joe Shaw - jshaw at
NetAdmin - Insync Internet Services
"Learn more, and you will never starve." - Paraphrase of Lee

I know this is slightly off topic, and I'll stop right here... Hope
everyone else has the sense to do the same.

On Thu, 26 Jun 1997, Joseph T. Klein wrote:

> Is the NAP.NET web server still a 90MHz Pentium running BSDI?
> Chris A. Icide wrote:
> > Interesting enough,
> >
> > Take this little tidbit.  The methodology in general was performance
> > testing of the backbone's web site.  (Which in itself is probably an
> > accurate measurement goal for the level of subscribers to that
> > magazine)  However, Nap.Net's web site was sitting behind a
> > Fractional T1 off of our chicago node while the web page was
> > being overhauled during this test period.  So, the rating for
> > Nap.Net reflects a Fractional (actually 22 channels) T1
> > performance.  This is interesting in the light that Nap.Net
> > still scored above the average and the median.
> >
> > Chris
> >
> > ----------
> > From:   Sean Donelan[SMTP:SEAN at SDG.DRA.COM]
> > Sent:   Thursday, June 26, 1997 5:34 PM
> > To:     nanog at
> > Subject:        Internet Backbone Index
> >
> > I guess it was only a matter of time.  The Keynote/Boardwatch
> > Internet Backbone Index
> >
> >
> >
> > I'm not going to comment on the methodology flaws, but it does
> > show what will fill the information gap.  I do like the line "The
> > world's fastest network provides good service to its users only if it
> > does a good job of connecting to the rest of the Internet."
> > Unfortunately
> > I don't think this study really shows that.
> > --
> > Sean Donelan, Data Research Associates, Inc, St. Louis, MO
> >   Affiliation given for identification not representation
> --
> Joseph T. Klein               |        mailto:jtk at
> Chief of Development          |

More information about the NANOG mailing list