Re: TCP (and SCTP) sucks on high speed networks

[Craig Partridge <craig@aland.bbn.com>]

In message <20001201221405.6F2C735DC2@smb.research.att.com>, "Steven M. Bellovi
n" writes:

>>Therefore, there needs to be a change to TCP's congestion avoidance algorithm
>>for future high speed networks.  Since SCTP is based on the same algorithms,
>>it is doomed to the same fate.
>
>...
>
>We, as a profession, don't know how to do better today, in the sense that 
>we don't have a solution that is (a) rationally deployable, (b) solves this 
>problem, (c) is "TCP-friendly", in the sense that it doesn't have undue 
>negative impact on existing RFC-compliant TCP streams, and (d) is 
>reasonably compliant with the Internet and the Internet philosophy.
>I'd love to see such a solution, but it is a research question.  We're 
>not going to solve it now, in the IPS working group.

I'd put it more firmly -- TCP's particular congestion algorithm is one
that has certain desirable stability properties -- any alternative needs
similar desirable stablity properties and we have very few such alternative
algorithms and, to my knowledge (and I'll confess I haven't spent time looking
into this literature deeply in recent years), none solves the problem of
fast restoration of full window after a packet loss without causing
potential congestion collapse.

Craig