[17:27:48] vkg joins the room [17:34:19] vkg leaves the room [17:45:05] Krishna Kalluri joins the room [17:45:12] Krishna Kalluri leaves the room [18:11:55] tomkri joins the room [18:38:24] vkg joins the room [18:40:57] chris_griffiths joins the room [18:48:06] Martin joins the room [18:48:40] Welcome to the ALTO WG jabber room. The today's agenda is here: http://www.ietf.org/proceedings/08nov/agenda/alto.html [18:49:02] slide are here:https://datatracker.ietf.org/meeting/73/materials.html [18:49:04] vkg leaves the room: Replaced by new connection [18:51:45] jan.seedorf joins the room [18:53:45] Martin leaves the room [18:54:03] Martin joins the room [18:54:43] saverio.niccolini joins the room [18:54:47] Krishna Kalluri joins the room [18:57:59] simon joins the room [18:59:34] Jim Galvin joins the room [19:01:30] saverio.niccolini leaves the room [19:03:32] Ted joins the room [19:03:48] slides are herehttps://datatracker.ietf.org/meeting/73/materials.html [19:04:03] now Status of the WG (http://www3.ietf.org/proceedings/08nov/slides/alto-6.ppt) [19:04:10] agenda bashing - accepted [19:04:24] Slide WG Status/History [19:04:32] Niklas joins the room [19:04:37] ludovic.poitou joins the room [19:04:42] saverio.niccolini joins the room [19:05:26] enrico joins the room [19:05:55] next slide ALTO WG [19:06:38] Who is presenting the slides? [19:06:40] jon-ietf joins the room [19:06:49] lisa dusseault, apps ad [19:06:59] Megumi joins the room [19:07:26] kivinen joins the room [19:09:12] resnick joins the room [19:09:26] Lars joins the room [19:10:02] Aaron Falk talking about p2p research group [19:10:09] no slides available [19:12:44] Leslie Daigle joins the room [19:12:54] Peny joins the room [19:12:59] see here for proposed updated p2prg charter: http://www.ietf.org/mail-archive/web/p2prg/current/msg01248.html [19:12:59] Enrico presenting the problem statement: http://www3.ietf.org/proceedings/08nov/slides/alto-3.ppt [19:14:05] slide 1 the problem [19:14:30] ray joins the room [19:15:08] next slide [19:15:12] changes from -02 [19:16:42] next slide [19:16:46] terminology [19:17:45] next slide: solution consider. [19:18:03] dmalas joins the room [19:19:45] next slide: [19:19:49] end of slides [19:21:15] Who is speaking? [19:21:17] jiang_xing_feng joins the room [19:21:22] vidja [19:21:23] Vidya Narayanan [19:21:42] Thanks [19:23:00] That was Ted Hardie/ [19:23:36] Sorry for that, thanks Pete [19:24:13] Sebastian presenting Requirements http://www3.ietf.org/proceedings/08nov/slides/alto-5.pdf [19:24:30] slide 1 with contributors [19:24:59] next slide presentation overview [19:25:14] next slide: External events.. [19:26:09] next slide: Changes [19:26:44] next slide: even more changes [19:26:45] Martin: More important for you to capture what is said that is not in the slides. [19:26:54] isn't there audio? [19:27:14] I can try to capture what is not on the slides [19:27:16] I think so; for those on the audio, the slide transitions are useful, at least in my past [19:27:27] next slide [19:27:33] Right. [19:28:47] It's easy to follow the audio when it's expanding on stuff on the slides. But Q's from the mic and side commentary it is easier to hear in the room than on the feed, and it's useful to have the jabber scribe write them down if he/she hears them. [19:28:48] next slide [19:28:59] overview of draft structure [19:29:21] ok - granted :) [19:29:30] will be a tough job - but I will give it a try [19:29:35] Thx! :-) [19:29:35] next slide [19:29:47] jabber scribe is always a tough job. ;-) [19:29:49] (Just trying to make your life difficult. :-) ) [19:30:00] hta joins the room [19:30:05] yeah :) [19:30:07] ....but, appreciated. [19:30:11] I'm in the room, BTW. Just trying to be helpul to listeners. [19:30:18] no problem [19:30:28] next slide [19:30:30] next steps(2) [19:31:26] end of slides [19:31:41] leslie daigle [19:31:46] speaking [19:31:55] some q about transport [19:32:03] xml over http considered [19:32:18] no requirements that imply specific solutions yet [19:32:31] Leslie: anticipated to have requirements about transport? [19:32:38] Vidya speaing [19:33:12] questioning the point of the discussion. [19:33:16] Simo Veikkolainen joins the room [19:33:19] John speaking [19:33:48] Vidya wants to have the backup slides discussed [19:34:11] quite a number of people have read the reqs document [19:34:51] Hannes speaking [19:35:05] time spent of reqs doc is a wasted time [19:35:17] some noise of discomfort about htis in the room [19:35:42] Hannes thinks group should spent time on real stuff [19:35:56] Ted Hardie [19:35:58] speaking [19:36:08] document still needs some work, too much implicit [19:36:26] shikob joins the room [19:37:43] Sebastian: assumed specific solution and generalized later [19:37:48] Dave Crocker [19:38:01] intersting topic for a decade w/o progress [19:38:32] concerns about talking in deltas [19:38:59] too much specifity und to little experience [19:40:02] [19:40:22] Hannes [19:40:44] looks like any other document no extra provided [19:41:19] saverio.niccolini leaves the room [19:41:19] resnick leaves the room [19:41:21] didn't get name -sorry [19:41:36] easy to make reqs document [19:41:50] resnick joins the room [19:42:02] The discussion is going out of topic [19:42:02] more difficult to define problem to solve [19:42:07] Glenn Zorn [19:42:09] tell me what the problem [19:42:11] thanks! [19:42:21] John speaking [19:42:26] requirements in the charter [19:42:33] person between crocker and hannes was matt mathis [19:42:53] Sorry it is Glen, not Glenn [19:43:06] Vijay speaking [19:43:24] ttfr joins the room [19:43:26] this here continued from p2pi workshop [19:43:53] Melinda joins the room [19:44:18] talk to people doing field tests and how to capture those [19:44:22] Joe Madlas [19:44:30] he likes req docs [19:44:41] helps to frame the work ahead [19:44:41] Ted leaves the room [19:44:52] Ted joins the room [19:45:30] Stas [19:45:56] ldondeti joins the room [19:46:34] John: do more discussions about this document [19:47:11] Next presentation is P4P Design and Implementation [19:47:16] http://www3.ietf.org/proceedings/08nov/slides/alto-2.ppt [19:48:06] next slide [19:48:10] P4P portal services [19:48:46] next slide [19:48:53] Location Portal Service [19:49:22] next slide: [19:49:24] PID [19:50:16] Niklas leaves the room [19:50:21] next slide [19:50:24] My Internet [19:51:07] omerboyaci joins the room [19:51:31] next slide [19:51:34] Implementation [19:52:41] next slide [19:52:45] pDistance Portal Service [19:52:59] question from the room [19:53:26] Ted Hardie [19:53:28] speaking [19:54:41] a particular AS was announce particular /16 subnet [19:54:54] portion of address space handed over to someone else [19:55:13] traffic engineering pratice [19:55:20] need to know hole punching rules [19:55:46] knowledge of full routing table [19:55:51] Anja Feldman speaking [19:55:59] who would be using this server [19:56:05] trackers and peers can use it [19:56:29] next slide [19:56:57] next slides [19:56:59] pDistance [19:58:07] next slide [19:58:28] Example: Routing cost pdistance of isp1 [19:58:34] next slide [19:59:03] next slide [19:59:11] Implem. Interface Defined... [19:59:18] Slide number will be really helpful [19:59:47] end of slides [19:59:51] jon-ietf leaves the room [19:59:52] there are no slide numbers [20:00:00] sorry but that's out of my control [20:00:13] people do not put the slide numbers in the slides [20:00:24] Ok No problem :) [20:00:40] omerboyaci leaves the room [20:00:45] Ted: Approach looking glass people to see whether you can get the same info [20:00:57] next presentation is Comcast's Experiences In a P4P Technical Trial [20:01:01] http://www3.ietf.org/proceedings/08nov/slides/alto-7.pdf [20:01:12] Rich is presenting [20:01:19] reports about testing done with p4p [20:01:37] is this the same p4p as was presented in the previous presentation? [20:01:57] yes [20:02:04] the first was p4p as tech [20:02:15] this is a result of a trail at comcast [20:02:33] next slide [20:02:35] key results [20:02:39] For those not familiar with looking glass, a quick summary is: [20:02:39] BGP Looking Glass servers are computers on the Internet running one of a variety of publicly available Looking Glass software implementations. A Looking Glass server (or LG server) is accessed remotely for the purpose of viewing routing info. Essentially, the server acts as a limited, read-only portal to routers of whatever organization is running the lg server. Typically, publicly accessible looking glass servers are run by ISPs or NOCs. [20:03:01] gacked from www.bgp4.as/lookinglasses [20:03:03] i'm puzzled by the tentative nature of the first presentation, then. if they had implemented it, I'd be expecting more specific opinions and implementation details. [20:03:58] they might have had more, but the presenter was running out of time. [20:04:03] next slide [20:04:07] p4p arch [20:05:28] next slide [20:05:33] comcast p4p trial parameters [20:05:47] Lars leaves the room [20:06:54] Barry Leiba joins the room [20:08:04] Lars joins the room [20:09:00] next slide [20:09:05] perf. summary [20:10:12] Heisenberg effect. The closer they measured it, the more they perturbed the results. [20:10:24] next slide [20:10:25] (yes, I'm joking) [20:10:29] comcast upload volumes [20:10:39] good joke :) [20:10:45] next slide [20:10:52] comcast download volumes [20:11:40] next slide [20:11:45] next steps ana dcaveats [20:13:17] vidya joins the room [20:13:35] Who is talking? [20:13:45] JP Vasseur [20:14:24] going back to requirements, I am curious who agrees with what is currently in the draft [20:14:48] at the moment, it is not clear if the requirements are for ALTO itself or for the apps using ALTO [20:14:55] Richard Yang speaking [20:15:07] omerboyaci joins the room [20:15:25] Vidya: How about asking this on the list, as this is the better place [20:15:37] now: ALTO Information Export Service [20:15:42] http://www3.ietf.org/proceedings/08nov/slides/alto-1.pdf [20:15:52] who is presenting? [20:15:58] first slide: protocol overview [20:16:01] did miss his name [20:16:08] sk joins the room [20:16:48] jiang_xing_feng leaves the room: Replaced by new connection [20:17:09] next slide [20:17:15] service operation (p2p example) [20:17:24] Martin: sure, I can take it to the list [20:17:29] thanks! [20:17:53] next slide [20:17:58] format of information [20:19:14] next slide [20:19:19] example usage by a p2p app [20:20:49] next slide [20:21:00] mapping ip addr to ASNs [20:21:10] saverio.niccolini joins the room [20:21:11] jon-ietf joins the room [20:21:27] hta -- What is your name re: you question about the implementation of P4P earlier? [20:21:42] next slide [20:21:47] other related issues [20:23:27] We've really tried to avoid mixing layers like that. Reusing the HTTP headers for this can cause you real confusion [20:24:04] Richard Barnes [20:24:07] speaking [20:24:22] check out work in cidr on mapping ip to asn [20:24:27] sidr [20:24:37] Anja [20:24:48] how much information clients are expected to download [20:25:24] Anja: didn't you want to download the routing tables [20:25:25] ? [20:25:59] dmalas leaves the room [20:26:04] only sends http get request to server [20:26:18] Lars leaves the room [20:26:58] dmalas: harald alvestrand [20:27:05] (hoping he reads the log later) [20:27:23] Stas: low effort [20:27:30] LEslie Daigle [20:28:32] jiang_xing_feng joins the room [20:28:37] Richard Woundy [20:28:58] look at p4p to get a rough figure about what data is exchanged [20:29:21] Hey, he left the v6 routing table out of the calculation [20:29:22] next presentation is Routing Proximity [20:29:28] http://www3.ietf.org/proceedings/08nov/slides/alto-0.ppt [20:29:31] Stefano presenting [20:29:36] next slide [20:29:38] intro [20:30:01] so -- a further point to my point: I'm not confident the colon-separated US-ASCII format would actually carry the desired range of IPv6 prefixes [20:30:04] for eg [20:30:26] next slide [20:30:29] arch [20:31:29] next slide [20:31:33] still "arch" [20:31:50] first bullet is :routing layer for proximity/ [20:32:48] Leslie Daigle leaves the room [20:33:22] Are there ospf equivalents to the isis attributes mentioned? [20:33:32] jon-ietf leaves the room: Replaced by new connection [20:33:37] jon-ietf joins the room [20:33:49] next slide [20:33:59] arch/standardization effort needed [20:34:37] [20:34:50] next slide [20:34:56] references - a lot of them [20:34:58] ;) [20:35:01] Leslie Daigle joins the room [20:35:24] John: is there dtraft [20:35:26] no. [20:35:32] Ruediger Volk [20:35:36] speaking [20:36:29] no reflection of spare capacatiy in routing sys [20:36:57] this stuff being realistic being entered [20:37:01] Stefano: [20:37:39] so far just investigation [20:37:47] Rich woundy speaking [20:38:01] routing topo changing over time [20:38:19] passing info to n clients for some time (hours, days, weeks) [20:38:32] smooth out version of routing info required [20:39:01] Stefano: clients shouldn't affected by immediate change but used in computing [20:39:16] hta leaves the room [20:39:45] (didn't understood his name, but he is with Alcatel-Lucent) [20:39:50] igor faynberg [20:39:57] thxs! [20:40:26] Barry Leiba leaves the room [20:41:06] Igor recommends to look at what the forces WG has done. [20:41:15] Richard Yang speaking [20:41:47] routing system assign metrics but peer selection is different problem [20:42:29] Stefano: believes in special config differing from today's routing info [20:43:17] Anja speaking [20:44:09] something about dynamic changes not being propagated properly, missing TE capabil.ities [20:44:50] next presentation [20:44:54] A Multi Dimensional Peer Selection Problem [20:45:00] http://www3.ietf.org/proceedings/08nov/slides/alto-4.ppt [20:45:24] Saumitra presenting [20:45:45] first slide: peer selection [20:45:56] Not able to open the presentation [20:46:27] works for me but is a .ppt file [20:46:40] next slide [20:46:46] inputs for... [20:47:13] Lars joins the room [20:47:42] next slide is [20:47:47] isp input not enough [20:48:33] next slide [20:48:36] overall desing gols [20:48:39] goals [20:49:37] next slide [20:49:45] multidimnesional alto [20:50:23] lellel joins the room [20:51:33] next slide [20:51:38] design recommendations [20:52:16] next slide [20:52:18] summary [20:52:40] I put a copy on my server: http://resnick1.qualcomm.com/ [http://resnick1.qualcomm.com/alto-4.pdf]alto-4.pdf [http://resnick1.qualcomm.com/alto-4.pdf] [20:52:45] end of slides [20:52:47] Leslie Daigle leaves the room [20:52:50] If it's still useful to anyone. [20:53:02] (In PDF since you were having trouble.) [20:53:02] Richard Woundy speaking [20:54:17] Stas speaking [20:54:24] saverio.niccolini leaves the room [20:54:24] resnick leaves the room [20:54:42] resnick joins the room [20:54:53] publishing infos might not be an incentive [20:55:18] Phil speaking [20:55:58] having interface but not describing the information provided at all [20:56:13] hard part is just deferred [20:57:30] Simo Veikkolainen leaves the room [20:57:59] resnick: Thanks for the PDF [20:58:12] WG to analyse what info to be exchanged [20:58:33] General comments to the WG [20:58:49] (missed name of speaker) [20:58:50] Melinda leaves the room [20:59:12] Ted leaves the room [21:00:11] session ended right now. [21:00:13] Jim Galvin leaves the room [21:00:13] bye bye [21:00:16] Martin leaves the room [21:00:16] shikob leaves the room [21:00:16] vidya leaves the room [21:00:19] jiang_xing_feng leaves the room [21:00:21] Megumi leaves the room [21:00:27] enrico leaves the room [21:00:30] ray leaves the room [21:00:37] Krishna Kalluri leaves the room [21:00:38] kivinen leaves the room [21:00:52] sk leaves the room [21:00:52] simon leaves the room [21:00:54] Lars leaves the room [21:01:08] jan.seedorf leaves the room [21:01:13] resnick leaves the room [21:01:47] tomkri leaves the room [21:02:36] chris_griffiths leaves the room [21:03:29] ludovic.poitou leaves the room [21:04:12] ttfr leaves the room [21:05:03] omerboyaci leaves the room [21:09:35] ldondeti leaves the room [21:14:46] dmalas joins the room [21:16:09] dmalas leaves the room [21:17:55] nordmark joins the room [21:18:27] nordmark leaves the room [21:19:02] dmalas joins the room [21:36:47] jon-ietf leaves the room [21:39:19] lellel leaves the room [21:52:30] simon joins the room [21:52:30] simon leaves the room [21:52:31] simon joins the room [22:04:53] simon leaves the room [22:37:15] Peny leaves the room [22:47:36] syi101 joins the room [22:47:44] syi101 leaves the room [22:50:06] dmalas leaves the room [22:51:57] dmalas joins the room