[ NNSquad ] Re: Cox to favor certain types of Internet traffic during "congestion"

["George Ou" <george_ou@lanarchitect.net>]

 [ I would categorize "protocol header analysis" as DPI.  

     -- Lauren Weinstein
        NNSquad Moderator ]


It would require nothing more than standard protocol header analysis and bit
counting to do this properly.  The actual payload never needs to be looked
at.

This classification is mostly true to the best practice priority
characteristics I laid out in my report in December
http://www.itif.org/index.php?id=205.

.	Low bandwidth and low duration (real-time applications like VoIP and
gaming)
.	Medium bandwidth and medium duration (streaming applications)
.	High bandwidth and low duration (Interactive applications like web
surfing and email)
.	High bandwidth and high duration (file transfer applications
including P2P, FTP, and web downloads)


This classification is justified by the fact that the end result is that
file transfer apps STILL end up with the highest average bandwidth despite
getting the lowest priority.


George

-----Original Message-----
From: nnsquad-bounces+george_ou=lanarchitect.net@nnsquad.org
[mailto:nnsquad-bounces+george_ou=lanarchitect.net@nnsquad.org] On Behalf Of
Lauren Weinstein
Sent: Wednesday, January 28, 2009 9:29 AM
To: nnsquad@nnsquad.org
Subject: [ NNSquad ] Cox to favor certain types of Internet traffic during
"congestion"



Cox to favor certain types of Internet traffic during "congestion"


( This would appear to *require* the use of DPI. )


http://www.washingtonpost.com/wp-dyn/content/article/2009/01/28/AR2009012801
003.html

http://www.cox.com/policy/congestionmanagement/

--Lauren--
NNSquad Moderator