Re: [RAM] Number of DFZ routers

Heiner,

> In einer eMail vom 07.07.2007 16:46:36 Westeuropäische Normalzeit 
> schreibt jnc@mercury.lcs.mit.edu:
> 
>       > A 200 k FIB should be proof enough that this is the wrong way to go.
>         > 2k should be sufficient for a 100 times bigger internet.
> 
>     Huh? The size of the FIB is caused by the way IPv4 addresses are
>     assigned,
>     and has nothing to do with the algorithm/protocol by which paths are
>     computed.
> 
>         Noel
> 
> Sorry, I was too sloppy. 2k, mayby 1k of addresses are even sufficient. 
> So I don't care about prefixes, neither PA nor PI, at all.
> But let's wait and see. There are other hurdles to take in the first place.

I completely agree with Noel. The FIB size mainly depends on how 
addresses are assigned and not on the routing protocol used to 
distribute the addresses. For a discussion on the FIB size with 
id/locator split solutions, you might have a look at :

B. Quoitin, L. Iannone, C. de Launois and O. Bonaventure.  Evaluating 
the Benefits of the Locator/Identifier Separation. Proceedings of 
MobiArch (ACM SIGCOMM Workshop), Kyoto, Japan, August 2007.
http://inl.info.ucl.ac.be/publications/evaluating-benefits-locator-identifier-separation-0

In this paper, we evaluate two possible scenarios for the assignement of 
locators in an id/locator separation solution and show that basically 
the max FIB size can be O(number of provider independant locator blocks).

Best regards,

Olivier Bonaventure

-- 
http://inl.info.ucl.ac.be , Universite catholique de Louvain, Belgium

_______________________________________________
RAM mailing list
RAM@iab.org
https://www1.ietf.org/mailman/listinfo/ram