[metrics-team] onionoo: significant improvement of AS level data coverage

[nusenu]

> I'd be interested to see (even as a one-off analysis) the
> agreement/disagreement between MaxMind's database and RIPE Stat.


They "disagree" about the ASes of 196 IP addresses (in onionoo data: 2018-03-19 13:00 UTC), 
but most are no real disagreements, they
just relate to IP prefixes belonging to multiple ASes and Maxmind only delivers one of them
(some big providers have multiple ASes and announce their prefixes on multiple ASes so there is no 
single truth about the IP -> AS mapping).

176 out of 196 are just caused by two organizations.


Maxmind failed to provide the AS number of two IPs.
RIPEstat also failed to provide the AS number of two IPs (not the same two as Maxmind).

In one case this is caused by RIPEstat's filters.
(RIPEstat filters routes with less than min 3 peers.)
https://stat.ripe.net/data/prefix-overview/data.json?resource=195.191.81.12
the second case is not clear to me
https://stat.ripe.net/data/prefix-overview/data.json?resource=77.68.213.162
https://bgp.he.net/ip/77.68.213.162



-- 
https://mastodon.social/@nusenu
twitter: @nusenu_

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.torproject.org/pipermail/metrics-team/attachments/20180319/21c72919/attachment.sig>