A couple of remarks to try and contribute to this discussion, one
philosophical and one engineering.
The philosophical first. There is a presentation taking place in a room
and it is possibly at q&a phase. You have several ppl entering the room
at several time points. What if every person entering the room,
immediately throws aggressively questions without even having spent time
to listen and understand the context of presentation and discussion.
Imagine this happening again and again. Consider what are the chances
that a meaningful communication and discussion will take place... or
simply how would you like that as a person in the audience?
The more technical next. Doubling and tripling IW for short lived
sessions (each of which will attain --only-- exponential growth in its
anyway short lifetime) means that a large number of high-freq transient
components will be added to and removed from the signal that the AQM
sees (and tries to adapt to). Is that something desired ? Will it
improve or worsen the way an AQM adapts? Increasing IW is not only a
matter of initial value, it also affects the starting growth rate (since
the relation is not linear).
Manos
On 25/09/14 19:49, Rick Jones wrote:
On 09/25/2014 10:26 AM, Mikael Abrahamsson wrote:
On Thu, 25 Sep 2014, Rick Jones wrote:
Well, there has been such a thing present in TCP from "The Beginning"
though not named as such. Such a client could always advertise a
smaller (initial) receive window... One which would allow only IW3 or
whatever value was appropriate.
I'm sure there are ways to solve this, but my take from the "TCP people"
was that there was not seen to be any need to do anything else than what
is done today, ie all TCP connections are self contained and learns
nothing from each other.
Well, I cannot speak for "TCP people" but I would think that what a
given TCP connection decides to advertise as its receive window, and
whether that decision would need/must depend on what other TCP
connections have seen are separate, but related.
The main point I wished to make was if one did indeed wish to have a
receiver behind a slow pipe wish to be able to keep a sender up on a
fast pipe from actually doing IW10, there was no need for any new signal
flowing from one end to the other, just setting the receive window
appropriately. The TCP stack on the slow-connection device could (and
perhaps should) be configured for things like (initial) receive windows
with that in mind.
rick jones
_______________________________________________
Bloat mailing list
[email protected]
https://lists.bufferbloat.net/listinfo/bloat
_______________________________________________
Bloat mailing list
[email protected]
https://lists.bufferbloat.net/listinfo/bloat
