As is our tradition by now a team of volunteers helped out with the network setup and operation of Free and Open-source Software Developers’ European Meeting (FOSDEM). The network was very similar to the one used over the last two years and we wanted to report on the evolution of the traffic we measured.

This year, we were able to go much ‘deeper’ into the traffic, so we have a lot more to report. Too much in fact for one article, so this is the first in a series.

First the configuration,  last year we faced performance issues when trying to capture netflow data, so now we used an ASR 1006 with RP2 and SIP20 modules. The configuration itself remained mainly the same with one difference: we changed the NAT64 setup to use the RFC 6052 well known NAT64 prefix of 64:ff9b::/96 which reduced the configuration on the ASR1006 to:

ipv6 access-list nat64-acl
 permit tcp 2001:67C:1810:F051::/64 64:FF9B::/64
 permit udp 2001:67C:1810:F051::/64 64:FF9B::/64
 permit icmp 2001:67C:1810:F051::/64 64:FF9B::/64
nat64 v4 pool nat64pool
nat64 v6v4 list nat64-acl pool nat64pool overload


We, however, retained much of the configuration, so like last year we counted the traffic on the interfaces via the interface counters and the number of packets using access-lists. First, we can see that the amount of data on the internet link increased significantly:


In fact, it almost doubled, and this is not only due to a large increase of the video streaming traffic, as we see a similar increase in the number of packets on the wireless client networks:


Besides this growth it’s also evident from the graphs that the IPv6 traffic (in brown) almost doubled.

If we look at the distribution between IPv4 and IPv6 traffic we notice that on the wireless network only 28% of the packets are IPv4, while on the internet side this increases to 67%. Possibly because of the NAT64 handling 6,139,812 translations from IPv6 to IPv4 only sites!

The first conclusions are clear: more users use IPv6 to access resources then last year and IPv6 traffic dominates on the wireless network.

However as I mentioned we collected a lot more statistics, I’m still working on the analysis. When we get more results we will continue this series. Next up: NBAR 2 results.