[Iccrg] Re: Heresy in Minneapolis
Bob Briscoe
rbriscoe at jungle.bt.co.uk
Tue Nov 18 14:25:20 GMT 2008
Lisong,
At 15:21 17/11/2008, Lisong Xu wrote:
>Bob,
>
>Thank for writing such an interesting and important draft!
>
>After reading it (quickly), it seems to me that the draft focuses on
>the fairness between different users. I agree that the fairness
>between different users is important. But I believe that TCP
>friendliness is proposed not only for the fairness between different
>users, but also for the fairness between different applications.
>
>So my question is while "volume accounting" helps on the fairness
>between different users (e.g. single connection vs multiple
>connections, active vs non-active), will it also help on the
>fairness between different applications (e.g. FTP vs VoIP,
>BitTorrent vs Joost)?
"Problem Statement: Transport Protocols Don't have to Do Fairness"
says both sides are wrong: neither TCP-friendliness nor volume
accounting. Please re-read.
However, that draft only describes the problem. My proposed answer to
your question about interaction between apps is 'congestion-volume'
accounting. This was described in another I-D that I allowed to
lapse, but published in CCR:
Flow Rate Fairness: Dismantling a Religion
<http://www.cs.ucl.ac.uk/staff/B.Briscoe/projects/refb/#rateFairDis>
To fully answer your question, a further paper (posted to this list
last week) describes a simple but unfamiliar token bucket mechanism
that we hoped would explain better how we think about fairness when
different applications interact.
Policing Freedom to Use the Internet Resource Pool
<http://www.cs.ucl.ac.uk/staff/B.Briscoe/projects/refb/#polfree>
Again, fairness here is a deeper level of fairness than flow rate.
It's down deeper - at the packet level - it involves integrating
mutual instantaneous congestion in all queues a user's traffic uses,
over time (easily measured as 'congestion volume').
more...
>By the way, we are also working on how to relax the traditional
>definition of TCP-friendliness. Currently, we are focusing on the
>impact of UDP-based applications on TCP-based applications. You can
>find more information here http://csce.unl.edu/~xu/research/stochasticTCP.html.
Yes, I already know your work. I'm afraid I don't agree. Please see
S.7.3 on TFRC in the CCR paper 'Flow Rate Fairness' above. We're in
the process of proving our arguments for a further paper due shortly.
As flows become less responsive to congestion on short timescales
(inelastic or semi-elastic), they are more likely to need to respond
at flow start (admission control). This is the motivation for the new
pre-congestion notification (PCN) standards, which the IETF's PCN
working group is just about to publish. <http://tools.ietf.org/wg/pcn/>.
Also relevant: see the minutes of yesterday's PWE3 session about
their congestion framework.
HTH
Bob
>Thanks!
>Lisong
>
>--
>Lisong Xu, Assistant Professor
>Computer Science & Engineering
>University of Nebraska-Lincoln
>http://cse.unl.edu/~xu
>
>Bob Briscoe wrote:
>>Matt & I have been having a side conversation about this.
>>In summary, we agree on what we don't want, but there's less
>>consensus on the path ahead.
>>I'd like to suggest that those interested in what the IETF needs to
>>do about relaxing TCP-friendliness make sure they're around in
>>ICCRG. I've bcc'd a few folks who I suspect will be interested but
>>might not naturally look in on ICCRG.
>>I haven't put this to the r-g chairs, but I imagine a discussion
>>will start in response to Matt's talk, and might need some time to
>>adjorn to a bar afterwards (there's one more session before the end
>>of the day afterwards tho). Perhaps we'll get together a truly ad
>>hoc Bar BoF :)
>>I know Matt is also talking on this in TSVWG & TSVAREA, but I
>>imagine ICCRG ought to be where any initial activity migrates to (&
>>I think Matt agrees).
>>My interest is that I believe TCP friendliness has become a
>>self-imposed barrier to innovation. What's the point of having the
>>e2e principle if you stop yourself and everyone else using the
>>freedom it gives on some dodgy grounds you can't really justify?
>>See "Problem Statement: Transport Protocols Don't Have To Do Fairness "
>><draft-briscoe-tsvwg-relax-fairness-01.txt>
>>
>>Bob
>>
>>At 03:23 16/04/2008, Matt Mathis wrote:
>>>On Tue, 15 Apr 2008, S. Keshav wrote:
>>>
>>>>Matt,
>>>> The paradigm holds sway only in the minds of academia. I
>>>> think (based almost purely on cynicism), that in the real world
>>>> TCP friendliness never had a chance. There is a long history of
>>>> TCP accelerators, multi-connection applications, UDP-blasters,
>>>> packet classifiers-and-delayers, and who knows what else that
>>>> have never cared about it. So, why do we need to phase it out?
>>>
>>>I would tend to agree. However, isn't this list supposed to
>>>represent academia?
>>>
>>>>Its already a done deal.
>>>
>>>Not in TCPM, TSVWG, etc, where dogmatic attachment to TCP-friendly
>>>is probably hurting the IETF. This is where we need to change
>>>minds and some deeply entrenched documents.
>>>
>>>I think the ADs are probably listening - do they have any commemts?
>>>
>>>Thanks,
>>>--MM--
>>>
>>>_______________________________________________
>>>Iccrg mailing list
>>>Iccrg at cs.ucl.ac.uk
>>>http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg
>>____________________________________________________________________________
>>Bob Briscoe, <bob.briscoe at bt.com> Networks Research Centre, BT Research
>>B54/77 Adastral Park,Martlesham Heath,Ipswich,IP5 3RE,UK. +44 1473 645196
>>_______________________________________________
>>Iccrg mailing list
>>Iccrg at cs.ucl.ac.uk
>>http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg
>
>
>_______________________________________________
>Iccrg mailing list
>Iccrg at cs.ucl.ac.uk
>http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg
____________________________________________________________________________
Bob Briscoe, <bob.briscoe at bt.com> Networks Research Centre, BT Research
B54/77 Adastral Park,Martlesham Heath,Ipswich,IP5 3RE,UK. +44 1473 645196
More information about the Iccrg
mailing list