[e2e] Open the floodgate

Cannara cannara at attglobal.net
Sun Apr 25 09:37:06 PDT 2004


Sam,

Apparently there's great bias against Source Quench because it requires a
packet be successfully sent in the reverse direction, but that's not such an
unreasonable thing anyway, given the congestion complained about is opposing.
TCP already provides that via a Window=0 statement back to the sender (which
is like what ECN can trigger).  

The reason 99% of congestion is handled by TCP's naive shutdown is, of course,
because it's so nonlinear, like Chicken Little.  If all flows experiencing
even the slightest loss go back to Slow Start, of course 'disaster' has been
averted.  This odd concept of a "control system" is like an automotive cruise
control that jams on the brakes when speed goes over the set point.  Internet
consumers, however, have much less understanding of how a network should
perform, so can easily be fooled by academic blab when they experience
slowness.

A really interesting insight was passed to me a while ago, which goes like
this:  Microsoft and the IETF actually have a synergistic relation -- a kind
of Yin & Yang of networking misunderstandings.  However adjustable TCP/IP
params may really be, Microsoft code behavior has instilled so much fear in
typical network managers, that whatever they may be able to change to improve
TCP in their environments they won't do, because it means touching the Dread
Registry.  And, you know what happens to people who try that.  :]

Alex


More information about the end2end-interest mailing list