I ran a test using TeliaSonera's Looking Glass Server and from their Los Angeles gateway, a traceroute to the v4 address of ntp1.home4u.ch is 11 hops away via UPC Broadband in Austria until it hits Cyberlink.ch's network in Switzerland.
In the v6 traceroute from the same source, it is 16 hops away and the path goes via CE Com Czech in The Czech Republic, ASCUS Telecom GmbH in Austria, before it reaches Cyberlink in Switzerland. So I would agree, v6 paths are likely much different than v4 paths. The reason is most likely that global v6 networks are much smaller with fewer links, and thus the paths between A and Z using v6 are fewer are farther between than in the v4 world which has been growing for the last 25+ years. I suppose a good way to think about it, is the difference between City Roads and Country Roads. In the city, there are likely hundreds of different routes that can be taken between two points. So finding the shortest path among all those options is pretty easy. In the country, there are fewer roads and many do not take you the most direct way to your destination. So if the only link between two points goes 20 miles out of your way, you are just out of luck until new roadways are put down. Same thing applies in the v4 vs. v6 realm. There aren't enough networks and interconnects handling v6 traffic, so the shortest path might be a long round-about way. v4: --- traceroute to 62.12.173.11 (62.12.173.11), 30 hops max, 40 byte packets 1 ash-bb1-link.telia.net (80.91.246.70) 69.819 ms ash-bb1-link.telia.net (213.248.80.9) 85.775 ms ash-bb1-link.telia.net (213.155.135.152) 62.055 ms 2 upc-ic-149905-ash-bb1.c.telia.net (213.248.85.146) 156.265 ms 206.106 ms 156.456 ms 3 us-was03a-rd1-xe-0-1-0.aorta.net (84.116.130.165) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 173.716 ms 170.650 ms us-was03a-rd1-xe-3-2-0.aorta.net (84.116.130.105) [AS 6830] 169.561 ms MPLS Label=300192 CoS=0 TTL=1 S=1 4 us-was02a-rd1-gi-6-0-0.aorta.net (84.116.130.157) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 177.236 ms 169.577 ms us-was02a-rd1-gi-9-0-0.aorta.net (84.116.130.153) [AS 6830] 196.854 ms MPLS Label=559 CoS=0 TTL=1 S=1 5 at-vie15a-ra1-xe-2-2-0.aorta.net (84.116.130.245) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 178.220 ms 169.698 ms 181.048 ms MPLS Label=300736 CoS=0 TTL=1 S=1 6 ch-zrh02a-ra1-xe-2-0-0.aorta.net (84.116.130.226) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 169.622 ms 169.952 ms 171.303 ms MPLS Label=305024 CoS=0 TTL=1 S=1 7 ch-zrh01b-ra1-ae-1.aorta.net (84.116.134.142) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 178.203 ms * 176.966 ms 8 62-179-117-174.ch-zrh01a-ra1.aorta.net (62.179.117.174) [AS 6830 <http://apps.db.ripe.net/search/query.html?searchtext=AS6830>] 171.696 ms 177.424 ms 188.264 ms 9 glbix-pe2-glbix-br1.cyberlink.ch (213.158.128.198) [AS 15623 <http://apps.db.ripe.net/search/query.html?searchtext=AS15623>] 200.459 ms 177.038 ms 174.697 ms 10 gw-fabianwenk.cyberlink.ch (212.55.218.77) [AS 15623 <http://apps.db.ripe.net/search/query.html?searchtext=AS15623>] 174.920 ms 188.250 ms 174.434 ms 11 ntp1.home4u.ch (62.12.173.11) [AS 15623 <http://apps.db.ripe.net/search/query.html?searchtext=AS15623>] 176.398 ms 193.993 ms 186.464 ms v6: --- traceroute6 to 2001:8a8:1005:1::11 (2001:8a8:1005:1::11) from 2001:2000:3018:2f::1, 64 hops max, 12 byte packets 1 s-b1-v6.telia.net (2001:2000:3018:23::1) 234.200 ms 183.135 ms 188.494 ms 2 interoute-ic-113261-s-b1.c.telia.net (2001:2000:3080:3e::2) 175.618 ms 190.509 ms 180.517 ms 3 2001:1478:0:1d::1 (2001:1478:0:1d::1) 262.231 ms 201.620 ms 202.804 ms 4 2001:1478:0:18::2 (2001:1478:0:18::2) 200.950 ms 209.276 ms 205.808 ms 5 2001:1478:0:b::1 (2001:1478:0:b::1) 209.131 ms 263.280 ms 216.753 ms 6 2001:1478:0:13::2 (2001:1478:0:13::2) 219.526 ms 220.208 ms 236.721 ms 7 2001:1478:0:f::2 (2001:1478:0:f::2) 217.951 ms 219.725 ms 216.148 ms 8 2001:1478:0:11::1 (2001:1478:0:11::1) 216.273 ms 226.218 ms 216.434 ms 9 2001:1478:0:12::2 (2001:1478:0:12::2) 216.715 ms 212.123 ms 217.609 ms 10 c76-3.ix1.vie.at.as39912.net (2001:1478:10:6::2) 217.289 ms 212.154 ms 215.765 ms 11 c76-1.ix2.ffm.de.as39912.net (2a01:f8:1:2:1:1:65:2) 228.951 ms 229.896 ms 229.287 ms 12 2a01:f8:2006:6::2 (2a01:f8:2006:6::2) 195.001 ms 327.708 ms 204.095 ms 13 glbix-br1-zhtix1-br1-via-a.cyberlink.ch (2001:8a8:1:1::35) 201.309 ms 199.299 ms 195.802 ms 14 glbix-pe2-glbix-br1.cyberlink.ch (2001:8a8:1:33::75) 199.392 ms 195.319 ms 203.697 ms 15 2001:8a8:0:185::100 (2001:8a8:0:185::100) 207.195 ms 198.819 ms 203.533 ms 16 ntp1.home4u.ch (2001:8a8:1005:1::11) 215.286 ms 214.144 ms 201.077 ms On Mon, Jun 18, 2012 at 8:08 AM, Fabian Wenk <[email protected]> wrote: > Hello Laurence > > On 17.06.2012 20:02, Laurence wrote: > >> I think I have the best looking IPv6 graph so far :) >> >> http://www.pool.ntp.org/user/**lskinner<http://www.pool.ntp.org/user/lskinner> >> >> The IPv6 box is stratum 1 with a GPS receiver. The IPv4 box sits next to >> it and is synced to it. Both go over the same VDSL FTTC line - the IPv6 >> is native, not tunnelled. IPv6 routing looks like it has some catching up >> to do to be as good as IPv4! >> > > I guess the IPv6 routing heavily depends on the peering / uplink from both > of the ISPs at each end. > > When I do traceroutes from me to your systems, I have 10 hops to your IPv6 > address, but 13 hops to your IPv4 address (output in the attachment). But > both path have around 55ms roundtrip time. > > As both our systems are located in Europe, and the NTP Pool monitoring is > located in L.A., it seems that the intercontinental routes can differ much > more with IPv6. > > It would probably be interesting so see the traceroutes from L.A. (the > monitoring system) to your IP addressess (IPv4 and IPv6) and also to my > system (ntp1.home4u.ch with IPv4 + IPv6). > > > bye > Fabian > > _______________________________________________ > pool mailing list > [email protected] > http://lists.ntp.org/listinfo/pool >
_______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
