I know 3 Geeks who lived in IPv4 15 port and IPv6 internet environment. They reported it was comfortable for home network.
And JANOG tested 15port of MAP-E. http://tools.ietf.org/html/draft-janog-softwire-report-01#section-4.9 MAP-E provides IPv6 to CPE,so CPE can use DNS resolver proxy. There is no performance impact,if they supported DNS proxy. Regards, -Shishio (2013/03/16 14:19), Xing Li wrote: > Satoru Matsushima 写道: >> As each implementation has its own, but I heard at least two implementors >> confirm that it is negligible difference between 15 to 63, done. > there is no performance impact for cernet's map cpe. xing >> cheers, >> --satoru >> >> On 2013/03/13, at 17:39, Satoru Matsushima <[email protected]> >> wrote: >> >>> Hi, I couldn't make comment during meeting but, >>> >>> On 2013/03/12, at 3:28, Ole Troan <[email protected]> wrote: >>> >>>> - Offset 4 versus 6. >>>> Discussion on value of wasting 3000 port versus simplicity of nibble >>>> alignment. >>>> No objection to moving default offset to 6 (from 4 in revision 04) >>> Does anyone have a evaluated result of NAT performance difference in the >>> number of derived port range? ie. 63 port ranges (6 bits offset) vs 15 port >>> ranges (4 bit offset) on a CPE. >>> >>> It would be better to see the result which show it could be negligible with >>> the updated text. >>> >>> cheers, >>> --satoru >> >> _______________________________________________ >> Softwires mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/softwires >> >> > > > > _______________________________________________ > Softwires mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/softwires _______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
