Vancouver IXP - VanTX - BCNet

Bill Woodcock woody at
Wed Aug 21 14:32:41 UTC 2013

On Aug 21, 2013, at 6:56 AM, Randy Bush <randy at> wrote:

>> Correct.  The ones in black are exchanges, the ones in gray are things
>> that someone asserted to have been exchanges, or asserted will be
>> exchanges.
> glad it's all so black and white, well grey. :)

When different people are asserting different things (i.e. that something is, and is not, an IXP) the situation is, by definition, contentious.  We move things into the "definitely an exchange" and show it in black text when we're able to observe a number of things:

 - Three or more participants
 - Shared layer-2 switch fabric across which participants peer with each other, exchanging customer routes
 - New participation is not too rigorously constrained (at least a domestic ISP new market entrant should be able to participate)
 - Participants do not receive a metered-rate bill based on utilization

In addition, we look for a number of signs of openness and transparency that would indicate that it's intended to be a good-faith effort to provide a point of interconnection between interested parties, rather than a regulatory compliance function, a set of private crossconnects that don't facilitation connection of new participants, a transit buyers'  club, or a commercial layer-1/2 WAN carrier trying to re-brand their services.  Which are, I would say, the four most common things that attempt to brand themselves as IXPs in disagreement with the general consensus that we observe.

New IXP founders typically contact our staff early in the formation process, and we collect the information above through conversations with participants, direct participation in the exchange, and on-site visits.  The weak point in this process is that when IXPs go defunct, we're often lacking a clear date of dissolution in our records, because they tend to fade away gradually, with very little public notice.

Whenever anyone notices such a discrepancy, we very much appreciate their bringing it to our attention, so we can make the directory more accurate.

> but how do you represent seattle colonolizing bc?

If, by that, you mean Canadian ISPs peering in Seattle, you'd see that in the participant list...

…and in theory the map here...

…should be showing the Canadian participation visually, but the fact that it's not, at the moment, is indicating a data coding error on our ARIN whois import, which I'll have our guys take a look at.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 841 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <>

More information about the NANOG mailing list