Re: [sunset4] Stateless NAT44 vs MAP/4rd

2012/7/20, Will Liu (Shucheng) <liushucheng@huawei.com>:
> Thanks for your interests in our draft.
>
> I am not quite sure whether you were asking the mapping between prefixs or
> IPs. Let me explain both. In our method, mapping between internal prefix and
> external prefix is 1:1, while mapping between internal IP and external IP is
> N:1. As shown in Figure 2 and 3, the length of external prefix is longer
> than the internal prefix. More specifically, the difference between them is
> the length of port set id. Did this answer the question, or do you have
> other questions regarding the mapping?

Thanks.
To be more precise, I guess N=2^(b-a) if I understood correctly
In another words, when network assigns external prefix, the
multiplexing ratio was been pre-determined by length of internal
prefix, i.e. deterministic NAT behaviour

BRs

Gang

> Regards,
> Will
>
>
>> -----Original Message-----
>> From: sunset4-bounces@ietf.org [mailto:sunset4-bounces@ietf.org] On
>> Behalf Of GangChen
>> Sent: Friday, July 20, 2012 2:02 PM
>> To: Simon Perreault
>> Cc: George, Wes; sunset4@ietf.org; Dan Wing; Will Liu (Shucheng)
>> Subject: Re: [sunset4] Stateless NAT44 vs MAP/4rd
>>
>> The draft is quite interesting
>> One question for clarification
>>
>> Is the mapping between Internal Prefix & External Prefix  1:1 or N:1 ?
>>
>> BRs
>>
>> Gang
>>
>> 2012/7/17, Simon Perreault <simon.perreault@viagenie.ca>:
>> > On 07/11/2012 12:44 PM, Dan Wing wrote:
>> >>> draft-tsou-stateless-nat44-00
>> >>
>> >> This seems to be a MAP solution.  But differs from MAP in that it uses
>> >> an IPv4 network between the subscriber and the ISP's "stateless NAT44"
>> >> device, whereas MAP uses an IPv6 network between the subscriber and
>> >> the ISP's "MAP border relay" device.
>> >
>> > I think this is a fair comparison.
>> >
>> > One advantage of SLNAT44 over MAP is that the necessary changes to the
>> > CPE are minimal. For example, a vanilla Linux can do SLNAT44 today.
>> >
>> > Another advantage of SLNAT44 is simpler provisioning: CPEs don't need
>> > to
>> > have access to all the "mapping rules" for direct communication to
>> > work.
>> > In MAP/4rd, a CPE needs to compute the IPv6 address to which the IPv4
>> > address it wants to reach is mapped. That's why rules need to be
>> > downloaded to the CPE. In SLNAT44, you just send the packet directly to
>> > the IPv4 address you want to reach. There is no overlay network, so no
>> > rules to be downloaded.
>> >
>> > Simplicity! :)
>> >
>> > Simon
>> > --
>> > DTN made easy, lean, and smart --> http://postellation.viagenie.ca
>> > NAT64/DNS64 open-source        --> http://ecdysis.viagenie.ca
>> > STUN/TURN server               --> http://numb.viagenie.ca
>> >
>> >
>> > _______________________________________________
>> > sunset4 mailing list
>> > sunset4@ietf.org
>> > https://www.ietf.org/mailman/listinfo/sunset4
>> >
>> _______________________________________________
>> sunset4 mailing list
>> sunset4@ietf.org
>> https://www.ietf.org/mailman/listinfo/sunset4
>