11:02 < Gergana_RIPENCC> Hello. Welcome to Connect WG. My name is Gergana from the RIPE NCC. I will be monitoring chat during this session. If you have a question or comment, I can read it out for you. Please make your questions as clear as possible. I will read out any questions when the speaker asks for questions at the end of his/her presentation. Please note that all chat transcripts will be archived and made available to the public on https://ripe77.ri[CUT]
11:03 < Gergana_RIPENCC> The Chair Remco van Took has started the session.
11:03 < Gergana_RIPENCC> Mook :)
11:04 < andyd> Best workgroup time o/
11:06 < Gergana_RIPENCC> George Nomikos has begun the presentation Uncovering Remote Peering Interconnections at IXPs.
11:09 < hlkv6> morning
11:11 < asbjorn> In our case (AS42541) we have our AMS-IX port directly to our AMS-CPH wavelength aka. DIY remote peering
11:12 < asbjorn> since all our customers are in CPH, it doesn't hurt any traffic
11:12 < ripe_561> k
11:13 < Gergana_RIPENCC> <asbjorn> Do you want me to read this out?
11:15 < asbjorn> Gergana_RIPENCC: sure, if he doesn't get to the point that it's not always harmful for routing
11:18 < ripe_561> Question for presenter : The work is motivated by the will to improve peering transparency about remote peers. Why is that so important to know if a network is remote or not, connected through a reseller or not ? Are there any risks linked to that ?
11:19 < ripe_561> The presentation seems to be biased against remote peering. But we see in this presentation this accounts for very little traffic and port utilisation, which suggests remote peering is used either for redundancy (not primarily) and / or as a test before becoming local. So is that so much of a bad thing ?
11:20 < andyd> Could the anonymous poster please indicate whether they really believe 40% of connections to the example large IXP given is connections for redundancy, or test connections?
11:20 < Gergana_RIPENCC> Thank you, I will read your comment out when the presenter asks for questions.
11:21 < asbjorn> if you want to peer with eg. netflix, you often have to be connected to an IXP far from your user base
11:22 < ripe_561> Other question for the presenter : As per previous works the methodology is based on RTTs. In the future IXP DB will gather member information from IXPs to populate the database. This will make it easier to find out who are connected members (locally or not) and who is connected via resellers. Will you adapt your methodology to make use of this tool ?
11:22 < andyd> Doesn't need reading out - but the slide showing 44% of paths using a remote path **when a shorter one exists** is our collective shame, and we deserve the broken internet that we are accidentally building.
11:23 < asbjorn> if we wanted to start accepting PNIs in AMS-IX, we would properly only put a switch in AMS, and still do routing in CPH
11:30 < anonymous_stoat> An IXP which is wide area is a sick, sick thing.
11:31 < Gergana_RIPENCC> <anonymous_stoat> Do you want me to read this as a comment?
11:31 < sbakker_as1200_ams-ix> Define "wide area"...
11:31 < sbakker_as1200_ams-ix> When is wide too wide
11:32 < anonymous_stoat> Gergana_RIPENCC: no.
11:46 < Gergana_RIPENCC> Andy Davidson has asked for questions.
11:50 < Gergana_RIPENCC> Nick Hilliard, INEX, has begun the presentation RPKI at Route Servers.
12:00 < Gergana_RIPENCC> Nick Hilliard has asked for questions.
12:01 < Apollo> Job throwing shade
12:13 < Gergana_RIPENCC> The presentation has ended.
12:13 < Gergana_RIPENCC> Florence Lavroff has begun the Connect Update.
12:16 < Gergana_RIPENCC> Bijal Sanghani has begun the Euro-IX Update.
12:19 < Gergana_RIPENCC> Thomas King, DE-CIX has begun his presentation on DE-CIX New York.
