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

["dave o'leary" <doleary@juniper.net>]

Thanks for the rant.

Have you considered the differences between a problems with a protocol
and the specific implementations of the protocol?  The adaptive congestion
response algorithms in TCP have been tweaked many times over the past
15+ years in a variety of research and commercial implementations.

                                         dave

At 11:44 PM 11/30/00 -0800, Matt Wakeley wrote:
>TCP's "congestion avoidance" algorithms are not compatible with high speed,
>long distance networks.  The "cut transmit rate in half on packet loss and
>increase the rate additively" algorithm will simply not work.
>
>Consider a 10Gbs link to a destination half way around the world.  A packet
>drop due to link errors (not congestion or infrastructure products) can be
>expected about every 20 seconds.  However, with a RTT of 100ms (not even
>across the continent), if a TCP connection is operating at 10Gbs, the packet
>drop (due to link error) will drop the rate to 5Gbs.  It will take 4 *MINUTES*
>for TCP to ramp back up to 10Gbps.
>
>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.
>
>-Matt