[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [DNA] Host BCP Issue 1: Same link vs Link change



Hi,

Some questions below...

<...>

>> I agree that DNA should detect link identity EFFICIENTLY, whether
>> link change or not.
>>
>> But I don't think that implies that a host needs to QUICKLY find out 
>> that it still remains at the same link.
>> In case of a link change, it's important for a host to quickly check 
>> for link change lest there should be service disruption.
>> But in case of no link change, I can't think of any harm from a little 
>> DNA delay.   

I am not sure I understand the above statement, so
here's the dumb question: how does a host know the link
has changed or not before finding out if the link has
changed or not?

I tend to agree with Sathya's suggestion:

 > FWIW, I think DNA should not differentiate between same-link and
 > link-change. DNA should define efficient mechanisms (in BCP and Soln.)
 > for the hosts to quickly identify its current link, whether it be the
 > same or different link.

This is of course from the host's bcp perspective.

Regards,

yushun
-- 
Yu-Shun Wang <yushunwa@isi.edu>  http://www.isi.edu/~yushunwa
USC Information Sciences Institute

S/MIME Cryptographic Signature