In einer eMail vom 12.01.2010 19:53:50 Westeuropäische Normalzeit schreibt  
[email protected]:

after  summarizing the main points, I'd also like to clarify the  
following  I caught from this thread of exchange:

> there are about 10 times  fewer active AS numbers than there are active
> prefixes. So  flat-routing on AS numbers would gain one order of  
>  magnitude
> immediately.

I've seen similar statement several  times by now.  I suppose people  
understand that routing on AS  numbers *alone* is not a feasible  
solution.
ASes come in all size  and colors.  Some AS could be a small company;  
some other AS  can cross multiple continents.  So AS does not represent  
a  granularity suited for today's routing needs. 
 

Routing  on ASes can  
get one reachability, but not performance, load balance,  traffic  
engineering etc etc.
Right.
By "no performance" I think you  mean the bad stretch factor  emphasized by 
Krioukov, applying to hierarchical routing concepts other than  TARA,
Traffic engineering: Indeed, only TARA provides a rearview mirror  
(enabling a communication between a congested node and just those who would use 
 it 
for transit) 
Moore's Law (only TARA would enable its applicability to IP  forwarding).
Mobility: TARA would help MIP4 enourmously and by the same token would do  
more for HIP than the HIP-origin RFC about rendex-vous server.
 
Heiner
 
PS: If I knew as much about the internet as you do, Lixia, I would  
certainly detect many more advantages of TARA :-)
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to