Reasonable tunnel performance?

[swsnyder]

I find that there's a ~60ms premium for going through the tunnel, relative to straight IPv4 traffic.

Code Select Expand


IPv4 ping of www.kame.net:

# ping -c 10 203.178.141.194
--- 203.178.141.194 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9002ms
rtt min/avg/max/mdev = 173.042/181.443/192.257/6.115 ms

IPv6 ping of www.kame.net:

ping6 -c 10 2001:200:0:8002:203:47ff:fea5:3085
--- 2001:200:0:8002:203:47ff:fea5:3085 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9000ms
rtt min/avg/max/mdev = 223.853/234.140/244.164/6.195 ms


FWIW, I'm going through the Chicago tunnel server.

Is this performance premium reasonable/typical for tunneled IPv6 traffic?

THanks.

[broquea]

The ipv4 path from your provider directly to the destination might not be the same path as sending the proto41 packets to our tunnel-server over v4 and then from there to the destination.

Trace to the tserv and the destination separately and you'll probably see a difference in path.