<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7655.10">
<TITLE>RE: Tunnel overhead [On killing IPv6 transition mechanisms]</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Brian<BR>
<BR>
Ping ietf with Ttl=66 waouw, I have never seen this value! I wonder which was the hop limit value at ietf web site :-)<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
Sent on my mobile with a micro keyboard. Please accept my apologies for typos...<BR>
<BR>
-----Original Message-----<BR>
From: Brian E Carpenter [<A HREF="mailto:brian.e.carpenter@gmail.com">mailto:brian.e.carpenter@gmail.com</A>]<BR>
Sent: Wednesday, March 17, 2010 01:41 AM W. Europe Standard Time<BR>
To: Gert Doering<BR>
Cc: ipv6-ops@lists.cluenet.de<BR>
Subject: Tunnel overhead [On killing IPv6 transition mechanisms]<BR>
<BR>
Gert,<BR>
<BR>
The problem is tunnels. The first example is from University of Auckland<BR>
via Telstraclear and a HE tunnel. IPv6 penalty is about 390 ms round trip,<BR>
crossing one ocean. The second is native all the way. *IPv4* penalty is 5 ms<BR>
round trip, crossing two oceans.<BR>
<BR>
The problem is not 6to4 specifically, but I don't have the ability<BR>
to compare 6to4 overhead with HE overhead.<BR>
<BR>
>ping www.ietf.org<BR>
<BR>
Pinging www.ietf.org [2001:1890:1112:1::20] with 32 bytes of data:<BR>
<BR>
Reply from 2001:1890:1112:1::20: time=667ms<BR>
Reply from 2001:1890:1112:1::20: time=544ms<BR>
Reply from 2001:1890:1112:1::20: time=575ms<BR>
Reply from 2001:1890:1112:1::20: time=344ms<BR>
<BR>
Ping statistics for 2001:1890:1112:1::20:<BR>
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),<BR>
Approximate round trip times in milli-seconds:<BR>
Minimum = 344ms, Maximum = 667ms, Average = 532ms<BR>
<BR>
>ping -4 www.ietf.org<BR>
<BR>
Pinging www.ietf.org [64.170.98.32] with 32 bytes of data:<BR>
<BR>
Reply from 64.170.98.32: bytes=32 time=141ms TTL=66<BR>
Reply from 64.170.98.32: bytes=32 time=150ms TTL=66<BR>
Reply from 64.170.98.32: bytes=32 time=142ms TTL=66<BR>
Reply from 64.170.98.32: bytes=32 time=142ms TTL=66<BR>
<BR>
Ping statistics for 64.170.98.32:<BR>
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),<BR>
Approximate round trip times in milli-seconds:<BR>
Minimum = 141ms, Maximum = 150ms, Average = 143ms<BR>
<BR>
----------------------------------------------------------<BR>
>ping www.surfnet.nl<BR>
<BR>
Pinging www.surfnet.nl [2001:610:1:80d0:194:171:26:201] with 32 bytes of data:<BR>
<BR>
Reply from 2001:610:1:80d0:194:171:26:201: time=311ms<BR>
Reply from 2001:610:1:80d0:194:171:26:201: time=310ms<BR>
Reply from 2001:610:1:80d0:194:171:26:201: time=311ms<BR>
Reply from 2001:610:1:80d0:194:171:26:201: time=310ms<BR>
<BR>
Ping statistics for 2001:610:1:80d0:194:171:26:201:<BR>
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),<BR>
Approximate round trip times in milli-seconds:<BR>
Minimum = 310ms, Maximum = 311ms, Average = 310ms<BR>
<BR>
>ping -4 www.surfnet.nl<BR>
<BR>
Pinging www.surfnet.nl [194.171.26.203] with 32 bytes of data:<BR>
<BR>
Reply from 194.171.26.203: bytes=32 time=322ms TTL=110<BR>
Reply from 194.171.26.203: bytes=32 time=313ms TTL=110<BR>
Reply from 194.171.26.203: bytes=32 time=313ms TTL=110<BR>
Reply from 194.171.26.203: bytes=32 time=313ms TTL=110<BR>
<BR>
Ping statistics for 194.171.26.203:<BR>
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),<BR>
Approximate round trip times in milli-seconds:<BR>
Minimum = 313ms, Maximum = 322ms, Average = 315ms<BR>
<BR>
<BR>
Regards<BR>
Brian Carpenter<BR>
<BR>
<BR>
On 2010-03-16 22:28, Gert Doering wrote:<BR>
> Hi,<BR>
><BR>
> On Tue, Mar 16, 2010 at 10:13:18AM +0100, Tore Anderson wrote:<BR>
>> on and on. Network latency is without question on that list. While<BR>
>> nothing can be done about a user's high-latency connection, 150ms of<BR>
>> additional client-server latency due to IPv6 being used will quickly add<BR>
>> up to seconds in terms of overall page load time for a complex web site,<BR>
><BR>
> I keep wondering about those 150ms. I seem to remember that Google saw<BR>
> *few users* that had a much higher IPv6 latency - but at the same time<BR>
> they also saw users with a *lower* latency via IPv6 (due to different<BR>
> network paths being taken).<BR>
><BR>
> For me, the v4/v6 latency is very similar - I use v6 all day, many of<BR>
> the sites I connect to (including google) have v6, and I do not see any<BR>
> adverse effects (nor any positive effects - it's just IP, after all).<BR>
><BR>
> So I'd like to see these 150ms qualified - in this discussion, it seems<BR>
> to be the assumption that "IPv6 will *always* be 150ms slower", which<BR>
> is most definitely not the case.<BR>
><BR>
> Gert Doering<BR>
> -- NetMaster<BR>
</FONT>
</P>
</BODY>
</HTML>