Iljitsch van Beijnum <iljitsch(_at_)muada(_dot_)com> writes:
I think there is some middle ground between 25000 and 10 ms.
10ms is the middle ground. That's enough for a bunch of retransmits on
modern hardware. Coupled with aggressive FEC, that's more than enough
time.
But the problem with sharing the airwaves is that you can't be
sure how long it's going to take to deliver packets.
Actually, the speed of light is remarkably deterministic. If the
network is so loaded that you can't send a packet in that period, you
should drop so that all the TCPs back off.
The difference between first try @ 11 Mbps and having to retry
several times @ 1 Mbps can easily be a factor 40.
None the less, it ends up being much lower by orders of magnitude than
what the standards currently permit.
The packet dumps I got from the 802.11b networks during the worst
periods at IETF revealed what you would readily expect -- that TCP
collapses badly when the underlying network does something very dumb.
By the way, it would also be a good idea if the standard did proper
power control of the mobile stations.
--
Perry E. Metzger perry(_at_)piermont(_dot_)com