[rrg] No liveness requirement in the ID/Loc Split concept

[Scott Brim]

Excerpts from marcelo bagnulo braun on Thu, Jan 15, 2009 10:19:28PM +0100:
> Scott Brim escribió:
>> Excerpts from Pekka Nikander on Thu, Jan 15, 2009 10:37:16PM +0200:
>>   
>>>> It's clear that pure endpoint-based multipathing, a la Shim6 REAP,
>>>> cannot scale.
>>>>       
>>> Would you please be more specific.  It is not at all clear to me.
>>> AFAIK, SCTP is doing essentially the same thing.  (I can see the
>>> signalling load there, causing potential packet storms; is that the
>>> "scalability" problem you are referring to?)
>>>     
>>
>> I should have done the math first but ... Let's assume,
>> conservatively, on the order of ten billion communicating endpoint
>> pairs and 2 locators per endpoint.  How often do you test locator
>> pairs that you are not using?  
> what do you mean by a "locator pairs that you are not using"?
>
> You mean a locator pair that is not the one that is the current locator  
> pair, or do you mean the current locator pair, but that you don't have  
> data packets to send in this moment?
> regards, marcelo

If one end has locators A and B, and the other end has C and D ...
then unless you are multipathing you will use a single pair, e.g. A-C,
while A-D and B-C are not being used, just checked for liveness in
case a need arises.