The revised version of the Faster Restart draft, available from "http://tools.ietf.org/html/draft-ietf-dccp-tfrc-faster-restart-04";, includes the following changes: Changes from draft-ietf-dccp-tfrc-faster-restart-03.txt: * Deleted ping packets, and the section about the implementation of ping packets in DCCP. * In Section 3.2, calls to "Update X_active_recv and X_fast_max;" and "Interpolate X_fast_max;" had been reversed accidentally. Put them back in the right order. * Changed Intended Status back to Experimental (where it started out). * General editing in response to feedback from Gorry. * Added simulation tests to the list in Section 6: (1) simulations with a worst-case scenario of high congestion, all flows using TFRC, all flows having various idle times, all flows using Faster Restart, and variable arrival rates for the TFRC flows (to create variable levels of congestion). And compare this to the same scenario with no flows using Faster Restart. (2) scenarios with transient changes from routing changes and from variable traffic. The goal is to explore worse-case scenarios showing off the worst aspects of Faster Restart. * Targeted an idle period of at most six minutes, not thirty minutes. Feedback from Gorry and Ian McDonald. * Added a section of whether Faster Restart encourages flows to pad their sending rate during idle periods. * Didn't implement suggestion from Lachlan Andrew to decay from qradupling to doubling the sending rate gradually. The last more-than-doubling of the sending rate is probably not a quadrupling in any case, since the allowed sending rate is not increased due to quadrupling to more than X_fast_max. - Sally http://www.icir.org/floyd/
