NNSquad - Network Neutrality Squad
[ NNSquad ] Re: Characterizing broadband networks without client software
|
I agree. What makes it especially interesting is the sense that
broadband is provisioned using traditional network/telephony thinking. I’m
trying to figure out how to articulate it but concepts like traffic shaping and
even MPLS would seem to bias the network towards a particular kind of traffic.
This is the insidious bias I’ve mentioned in the past. With the
descriptions of the queue delays you get a sense that managing scarce resources
that is deeply embedded and implicit in the psyche of these network planners. You
also get a sense of the consequences of building on top of tradition networks
as in the delays waiting for your slot in the TDM stream. If you look at 3G cellular architecture (there was a fine
issue of IEEE Communications on EVDO a while back) you see similar the policies
built in below the IP layer that limit the use of the network as simple
infrastructure. -----Original Message----- Wes Felter <wesley@felter.org> > This paper appeared in this year's Internet Measurement
Conference: > Marcel Dischinger, Andreas Haeberlen, Krishna P.
Gummadi, Stefan > Saroiu: Characterizing Residential Broadband
Networks. > http://www.imconf.net/imc-2007/papers/imc137.pdf This very readable paper is a must see. I urge all
NNSquad participants to give it a look ASAP. Thanks, Wes! --Lauren-- NNSquad Moderator |