> From: Paul Jakma <[email protected]>
> I'm not sure it's enough just to try several. That will catch
> 'workability' problems, as you suggest, but it won't per se tell us
> which solutions handle growth the best, will it?
Well, nothing will really, _absolutely_, tell you except full-size
deployment, neh? (As the electrial grid guys have found to their cost, on
several occasions...) Simulations, good engineering practices, etc all help,
but in the end...
> it'd be really useful to model some of the different options against
> each other. If that hasn't been done already? (I'm presuming not).
Ah, actually, it has. Here are the latest results from modelling of
aspects (mapping lookup time) of one
scheme:
http://www.ietf.org/mail-archive/web/lisp/current/pdf1piLVmIlgK.pdf
and here is another paper which covers an earlier round of modelling (on
cache size, hit ratios, etc):
http://inl.info.ucl.ac.be/system/files/Conext-2007-CRV-UCL-v2-Clean.pdf
I'm pretty sure there's been some ILNP modelling, but I don't recall for
sure; try looking on the ILNP page:
http://ilnp.cs.st-andrews.ac.uk/
I don't see anything there which says 'modeling' in the title, but I didn't
have time to look at them all.
I don't know if any of the others have been modelled; I looked on the RRG
page:
http://trac.tools.ietf.org/group/irtf/trac/wiki/RoutingResearchGroup
but I didn't see anything else likely looking.
> Might there be some pretty rich seams here to mine for academic work?
Already happening... :-)
Noel
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
