[e2e] TCP and MAC layer

Harald Alvestrand harald at alvestrand.no
Wed Oct 31 00:28:06 PST 2001


to make the opposing view:

in the 95% case, there is a strong resembleance between my current 
connectivity and the connectivity of 1 minute ago.
Core route changes generally switch between same-speed paths (and when they 
don't, the fallback path is so congested as to be useless anyway); 
intra-technology roaming often will not produce drastic capacity/latency 
changes; inter-technology roaming doesn't happen that often (on the order 
of minutes to hours, not seconds, even if we can make it work seamlessly, 
which we mostly can't today).

Guessing "right" for initial RTTE and window size 95% of the time is 
potentially considerably better than what you get with a fixed initial 
guess.

Just being heretical....

                  Harald


--On tirsdag, oktober 30, 2001 11:20:51 -0800 Joe Touch <touch at ISI.EDU> 
wrote:

>> I suspect knowing the latency of the path is more useful than the
>> next-hop link capacity in this instance. High latency -> larger
>> initial window to reduce round-trip time opening window. The question
>> is whether you can safely reuse such knowledge from previous
>> connections, and generally you can't.
>
>
> Whether you can or not depends on whether you have data that is a
> reasonable predictor. For the average net-connected system, this may not
> be true, but for systems on the terminus of a satellite link, this may be
> just fine.





More information about the end2end-interest mailing list