[Iccrg] Aggregate congestion control

[Michael Welzl]

Hi all!

Ideas for aggregate congestion control tend to pop up every now and then 
- e.g. there's an old paper on doing this as a defense against DoS at:
http://ccr.sigcomm.org/archive/ccr-toc/ccr-toc-2002.html (vol 32 no 3, 
search for "aggregates" on the page)

And there were papers that used TFRC (turned into "CP"):
D. E. Ott, T. Sparks, and K. Mayer-Patel. Aggregate Congestion Control 
for Distributed Multimedia Applications. In IEEE Infocom 2004, Hong Kong.
...and MulTCP (turned into "PA-MulTCP"):
F.-C. Kuo and X. Fu. Probe-Aided MulTCP: an aggregate congestion control 
mechanism. SIGCOMM Comput. Commun. Rev., 38(1):17–28, 2008.

This work focuses on traffic engineering, but is also about controlling 
aggregates in a way:
http://nms.csail.mit.edu/projects/texcp/

And this paper, for instance, introduces an XCP-based PEP for a path 
segment, to improve satellite communications:
http://www.isi.edu/isi-xcp/docs/kapoor-pep-gi2005.final.pdf
..., including the statement: "There is a one-to-one mapping between TCP 
and PEP connections. We did not implement multiplexing of TCP 
connections, mainly for simplicity. It remains a promising direction for 
future work."


There's got to be more that I don't know about (and I'd be thankful for 
pointers). But I'm mostly interested in the general perception - is 
aggregate-based congestion control regarded as a promising direction? 
(why?) As futile? (why?) Is anybody actually deploying an aggregate 
congestion control based solution? I'd just like to hear some thoughts.

Cheers,
Michael