I am sorry that I was not able to be at IETF last week.
This email is in response to the discussion of the slowly-responding
variant of CCID2 in the minutes.
The internet-draft:
There is a very rough draft on this, from December 1, 2006, available
from
"http://www.icir.org/floyd/papers/draft-floyd-dccp-ccid2slow-00b.txt";.
It is written as a variant of CCID 2, defined simply by adding a
new AIMD Category feature for CCID 2. That is, it is intended to be
something simple and straightforward.
Flow-rate fairness and other issues:
There is a fair amount of research comparing TFRC with TCP-like
congestion control with slowly-responding parameters. One such
paper is the following:
FLOYD, S. AND HANDLEY, M. AND PADHYE, J. A Comparison of
Equation-Based and AIMD Congestion Control, May 2000.
Available from http://www.aciri.org/tfrc/.
Related literature includes the following:
BANSAL, D., AND BALAKRISHNAN, H. Binomial Congestion
Control Algorithms. In Proceedings of the Conference on Computer
Communications (IEEE Infocom) (Anchorage, AK, April 2001),
pp. 631640.
CHIU, D.-M., AND JAIN, R. Analysis of the Increase and Decrease
Algorithms for Congestion Avoidance in Computer Networks.
Computer Networks and ISDN Systems 17 (1989), 114.
YANG, Y., AND LAM, S. General AIMD Congestion Control. Tech.
Rep. TR-2000-09, University of Texas at Austin, May 2000.
Available from http://www.cs.utexas.edu/users/lam/
NRL/TechReports/.
This is not what you would call new territory...
- Sally
http://www.icir.org/floyd/
The caveat:
I am slow to get to things these days, and work at most half time,
and have eight current internet drafts on my plate, in addition to
the thirteen internet-drafts-to-be, including this one, listed on
my todo list at "http://www.icir.org/floyd/papers.html";. So if
anyone was interested and wanted to help move this along, I think
that would be great. If not, that is fine too, and this will proceed
at whatever pace it proceeds at. That is, there are no promises.
Just FYI.
