>> On Sun, 9 Oct 2011, Shishio Tsuchiya wrote:
>>
>>> 2^64 host address space would be too enough for homenet.
>>
>>
>> I would just like to comment on this even though you retracted your question.
>>
>> As an ISP, I do not want to participate in a home with hundreds or thousands
of active IPv6 addresses that I need to keep state for. Most likely, when we
have central equipment participating in a home network, we'll implement a limit
on number of neighbours it'll allow on the subnet. Doing ND/NS a lot takes
resources.
>>
>> If customer wants to use more IPv6 addresses, we'll require them to get a
home router and PD address space to it to distribute the ND/NS load.
>>

>Agreed. But I think it is worth noting that nd-proxy is  practical in mobile
networks where each subscriber gets a unique /64.

>Mobile network attachment has popped up a few times in homenet discussions.


Yes, this will useful in a number of use cases with mobile attachment.
Tethering and other like operations come to mind.
A good reference is - draft-ietf-v6ops-3gpp-eps for mobile operation.

Victor K


>Cb
>> -- 
>> Mikael Abrahamsson    email: [email protected]
>>
>> _______________________________________________
>> homenet mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/homenet
_______________________________________________ homenet mailing list
[email protected] https://www.ietf.org/mailman/listinfo/homenet

_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to