[Iccrg] Re: [Tsvwg] review solicitation for ICCRG congestion
control survey
Michael Welzl
michael.welzl at uibk.ac.at
Fri Jan 12 07:19:28 GMT 2007
Hi Mitchell,
Thanks for your comments!
> Doesn't this document initially imply that ONLY
> "feedback-based adjustment" type responses are available
> to ameloriate congestion.
>
> As congestion approaches, the lack of acks/feedback can
> also indicate congestion. So I would first modify section
> "1" to indicate that the loss of feedback is also a level
> of feedback that needs to be considered.
Agreed. Indeed, the lack of feedback is a form of
feedback itself.
In the remainder of your email, you raised some interesting
research questions - but I don't think that our document
should address these, as our intention is not to provide
a discussion of all the design pro's and con's of
congestion control (which is an almost endless story).
For example:
> In addition, this reader is unaware whether this document prefers
> flows that are at the precipice of congestion or whether a
> significant buffer of lost bandwidth should be used to minimize
> any congestion type behaviours.
the draft doesn't prefer anything; it is merely
meant as a survey of what the IETF has done in
this area.
Of course, no such document can stand without some
introductory statements about the general field
that is being addressed - and therefore I fully
agree with your correction regarding section 1.
Cheers,
Michael
More information about the Iccrg
mailing list