Re: [v4tov6transition] Any Experience with Using Behave's

[Francis Dupont <Francis.Dupont@fdupont.fr>]

StatelessNAT-PT for IMS-SIP VoIP Application...
In-reply-to: Your message of Wed, 22 Sep 2010 09:23:09 PDT.
             <AANLkTi=+Fx4c6UvcBi3GO2U2Cg9y7EkkTcmyO2q_YQ6t@mail.gmail.com>
Date: Wed, 22 Sep 2010 21:53:55 +0200
Sender: owner-v6ops@ops.ietf.org
Precedence: bulk
List-ID: <v6ops.ops.ietf.org>

 In your previous mail you wrote:

   Generally about NAT64 performance, we expect it to be approximately
   consistent with NAT44 CGN performance, slightly less on some
   platforms.  ALGs generally decrease performance since they require
   more complex logic deeper in the packet.

=> as a writer of both (NAT-PT and CGN) packet processing in protocol
translation is clearly more complex and there are things like checksums
or (brrrr) fragment handling which are more expensive.
Now the time is spent into packet to NAT entry structure lookup for,
(best algorithms are in O(log(number_of_entries))) so as soon as the
NATxx is heavily loaded this will be the same.
BTW I fully agree about ALGs (and the worst case is SIP/SDP/etc).

Regards

Francis.Dupont@fdupont.fr