From: Ryan Anderson <randerso@ece.eng.wayne.edu>
Date: Sat, 12 Jul 1997 16:57:33 +0800
To: Steve Schear <azur@netcom.com>
Subject: Re: Routing around damage
In-Reply-To: <v03102802afecd9bc6add@[10.0.2.15]>
Message-ID: <Pine.GSO.3.95.970712041915.26608B-100000@ece>
MIME-Version: 1.0
Content-Type: text/plain



On Fri, 11 Jul 1997, Steve Schear wrote:

> >> Remember "the internet interprets censorship as damage and routes around
> >> it"?  I'd be happy with an internet that interprets DAMAGE as damage and
> >> routes around it.
> >
> >It does.. It's just that when you lose a *large* access point, the impact
> >is significant.  (I think that's what happened here...)
> 
> Seems to me that having only a few, heavily trafficed, NAPs is a
> topological weakness in the Net which needs to be delt with soon.

Quite true, but the other main problem, that of major routers dropping
lots of packets during high traffic times (basically, anytime from noon
eastern until about 8 eastern is horrible, and it tends to pick up a
little after that I think, but it stays bad until after about midnight
eastern from what I've been able to tell), remains completely unresolved.
Some initiatives are in the planning stages to deal with these things, but
the sheer number of routes on the net at this point is a problem.  Of
course, MCI's horrible ATM <-> IP interface is a major point of congestion
at this point, at least from my connection here at school.

Ryan Anderson - <Pug Majere>     "Who knows, even the horse might sing" 
Wayne State University - CULMA   "May you live in interesting times.."
randerso@ece.eng.wayne.edu                        Ohio = VYI of the USA 
PGP Fingerprint - 7E 8E C6 54 96 AC D9 57  E4 F8 AE 9C 10 7E 78 C9
-----------------------------------------------------------------------