[Iccrg] Fwd: agenda planning for ICCRG in Anaheim

[Michael Welzl]

On Feb 22, 2010, at 12:08 AM, Lachlan Andrew wrote:

> Greetings Michael
>
> On 22 February 2010 09:07, Michael Welzl <michawe at ifi.uio.no> wrote:
>> On Feb 21, 2010, at 10:48 PM, Matt Mathis wrote:
>>>
>>> Therefor: all stacks should include a secondary congestion control
>>> mechanism that detects the delay caused by large queues in the  
>>> network
>>> and regulates their congestion window accordingly.
>>
>>> It may be an appropriate future work item of the ICCRG and IETF to
>>> attempt to standardize such a mechanism.   In the short term it  
>>> would
>>> be sufficient to merely agree on a statement of principle or intent.
>>
>> - attempting to standardize: isn't that exactly what LEDBAT is doing?
>
> I thought LEDBAT was specifically for less-than-best-effort.  Matt's
> proposal is that *all* stacks should consider delay.  That is, we also
> want a replacement which is competitive with standard TCP "on
> average", although perhaps less aggressive when buffers are large and
> more aggressive either in the presence of non-congestion loss or when
> recovering from transient congestion.

I just said that LEDBAT is attempting to standardize a mechanism
that does such a thing (no matter what its focus is). I didn't interpret
Matt's statement to mean that we also want a replacement which is
"on average" competitive with standard TCP, but such a mechanism
is surely an interesting consideration.


> CTCP fits the final part of that (recovering faster from transient
> congestion), but misses the other components.
>
> Also, since LEDBAT is IETF not IRTF, it is not doing research.  It
> will standardise something, but probably not the best thing.  The
> ICCRG can take a longer term view, and hopefully get it right before
> putting a proposal to the IETF.

Agree 100%, and I like the requirements you lay out above
for a delay-based mechanism. I find it a bit hard to imagine
what such a mechanism could look like, though, but where
there's a will, there's a way, they say...

Cheers,
Michael