IETF
icnrg@jabber.ietf.org
Thursday, July 28, 2022< ^ >
Room Configuration
Room Occupants

GMT+0
[13:59:04] <zulipbot> (Colin Perkins) I think it auto-imports what's in the datatracker when the session opens
[14:18:30] <zulipbot> (Lixia Zhang) I believe the very initial idea (of using an Interest to set up path for another interest) came from Yu Zhang when he visited UCLA. Wonder whether the IPR application ACK'ed that fact.
[14:21:15] <zulipbot> (David Oran) Well, for 3rd party declarations you only say what you know about, not what others may know about. It would be up to Yu Zhang to put one in. Is there anything written down? I'd be interested
[14:22:12] <zulipbot> (Dirk Kutscher) Shouldn't the camera shift to Thomas (active speaker)?
[14:22:35] <zulipbot> (Dirk Kutscher) thanks!
[14:23:01] <zulipbot> (Lixia Zhang) @DaveO: yes Yu Zhang has an ICN poster some years back, also an NDN Tech report at the time.
[14:26:10] <zulipbot> (Carsten Bormann) It is pretty easy to pick apart a float32 and get this
[14:33:17] <zulipbot> (David Oran) @Lixia - thanks I'll try to dig this up - the stuff in the draft comes from Ilya & my ICN'17 path switching paper, so the patent application predates that by a few months or so.
[14:36:50] <zulipbot> (Lixia Zhang) I just dug out the poster info:
Poster: "Kite: A Mobility Support Scheme for NDN"
Yu Zhang, Hongli Zhang, Lixia Zhang
ACM Information Centric Networking Conference, 2014
[14:49:55] <zulipbot> (David Oran) My incoming audio is nearly unintelligible - meetecho saying that the audio stream is only 8-9 kbps, which is clearly not enough. Is there any way to force a higher bitrate inbound?
[14:50:08] <zulipbot> (Eve Schooler) i too am having on again off again audio issues
[14:50:34] <zulipbot> (Carsten Bormann) I get 16 kbit/s, had a brief period of high loss
[14:51:14] <zulipbot> (Carsten Bormann) Oh, and the videos break away now and then
[14:52:06] <zulipbot> (David Oran) The video is jerky too, but turning off the video feeds does NOT improve the audio, and i have plenty of incoming bandwidth to the machine..
[14:52:32] <zulipbot> (Carsten Bormann) audio is near-perfect now.
[14:53:05] <zulipbot> (Lorenzo Miniero) We're aware of this kind of issues for some people, which may be related to some routing issues: we're investigating with the NOC
[15:02:48] <zulipbot> (David Oran) So, a speedtest from my machine shows 600+MPBS download, so bandwidth isn't the problem. And audio is terrible with two different browsers (safari & chrome)
[15:03:14] <zulipbot> (David Oran) Any suggestions?
[15:03:27] <zulipbot> (Lorenzo Miniero) @**David Oran** yes, as I wrote above it's likely not a user problem but an issue somewhere else, we're looking into this
[15:04:17] <zulipbot> (David Oran) @lorenzo - thanks - letme know if there's anything yiu want me to try
[15:04:17] <zulipbot> (Lorenzo Miniero) Did you try refreshing the page already?
[15:04:30] <zulipbot> (Carsten Bormann) Dave: traceroute www.conf.meetecho.com
[15:04:43] <zulipbot> (David Oran) @lorenzo - yes, multiple times, plus leaving and rejoining, and switching browsers.
[15:04:56] <zulipbot> (Lorenzo Miniero) Ack, was worth a try :)
[15:05:22] <zulipbot> (Lorenzo Miniero) Carsten its wws, not www
[15:06:14] <zulipbot> (David Oran) @lorenzo - I'm going to kill the browser completely and restart, just for fun. Should i try Safari, Chrome, or Firefox?
[15:06:15] <zulipbot> (Carsten Bormann) Ha, that's what my brain wanted to type, but my fingers didn't obey
[15:06:56] <zulipbot> (Lorenzo Miniero) All browsers should be ok, even though Safari is known to be a bit more picky at times :)
[15:07:22] <zulipbot> (Carsten Bormann) Now lost the chat in the meetecho window
[15:07:36] <zulipbot> (Dirk Kutscher) I'm using Brave (Chromium-based), and there are no issues.
[15:10:59] <zulipbot> (Carsten Bormann) It is usually easier with inexperienced presenters if the chair picks the slides and then hands over control
[15:11:18] <zulipbot> (David Oran) @lorenzo - identical bad audio with a nice fresh Firebfox instance.
[15:11:46] <zulipbot> (Colin Perkins) Right - passing control of the slides works very well, in my experience
[15:11:59] <zulipbot> (David Oran) @Lixia: Oh, I know about Kite. It doesn't force packets onto a path, it just ensures the forwarder state is there.
[15:12:55] <zulipbot> (Dirk Kutscher) @Colin, thanks!
[15:13:08] <zulipbot> (Lorenzo Miniero) @**David Oran** yeah if it's indeed a network issue somewhere changing browser won't help unfortunately. I'm waiting for an update so I'll let you know when I know more. As Carsten suggested, you can check the path between your location and the server
[15:15:02] <zulipbot> (David Oran) @carsten, @lorenzo % traceroute wws.conf.meetecho.com
traceroute: Warning: wws.conf.meetecho.com has multiple addresses; using 35.156.222.202
traceroute to meetecho-lb-new-1290277547.eu-central-1.elb.amazonaws.com (35.156.222.202), 64 hops max, 52 byte packets
1  192.168.15.1 (192.168.15.1)  3.594 ms  0.467 ms  0.402 ms
2  96.120.66.93 (96.120.66.93)  12.818 ms  10.490 ms  8.087 ms
3  24.124.212.245 (24.124.212.245)  13.658 ms  11.653 ms  9.098 ms
4  96.110.22.85 (96.110.22.85)  10.197 ms  10.741 ms  13.138 ms
5  be-1003-pe02.onesummer.ma.ibone.comcast.net (68.86.90.173)  13.420 ms  11.813 ms  12.625 ms
6  50.208.233.110 (50.208.233.110)  10.711 ms  15.196 ms  11.801 ms
7  ae21.cr10-lon1.ip4.gtt.net (89.149.139.5)  86.962 ms  96.132 ms *
8  ip4.gtt.net (154.14.130.134)  85.596 ms  89.016 ms  88.275 ms
9  * * *
10  * * *
11  * *^C
[15:15:29] <zulipbot> (David Oran) traceroute6 wws.conf.meetecho.com
traceroute6: Warning: meetecho-lb-new-1290277547.eu-central-1.elb.amazonaws.com has multiple addresses; using 2a05:d014:2d4:4101:96f0:9bc3:7887:6f3d
traceroute6 to meetecho-lb-new-1290277547.eu-central-1.elb.amazonaws.com (2a05:d014:2d4:4101:96f0:9bc3:7887:6f3d) from 2601:184:407f:80cf:7c55:6c59:e591:202b, 64 hops max, 12 byte packets
1  2601:184:407f:80cf::1  2.738 ms  0.973 ms  0.910 ms
2  2001:558:4023:44::1  9.278 ms  10.691 ms  10.661 ms
3  2001:558:202:4065::1  11.485 ms  11.849 ms  10.082 ms
4  * * *
5  * *
[15:16:29] <zulipbot> (David Oran) ok, the audio just resolved right after doing the traceoutes. Huh???!!!
[15:18:12] <zulipbot> (Dirk Kutscher) Just noting: I'm connecting via IPv4 (not issues).
[15:18:25] <zulipbot> (Dirk Kutscher) (no issues)
[15:21:20] <zulipbot> (Carsten Bormann) I get the AWS server at hop 15; hop 6 to 14 don't reply.
In Europe, BTW, ec2-3-67-108-167.eu-central-1.compute.amazonaws.com
[15:24:12] <zulipbot> (Dirk Kutscher) Something else: there is a significant delay when sending a chat message (before you see it in the list). Is that just me?
[15:25:05] <zulipbot> (David Oran) I see it too
[15:25:32] <zulipbot> (Dirk Kutscher) It's +10 seconds...
[15:25:45] <zulipbot> (David Oran) ditto
[15:26:59] <zulipbot> (Colin Perkins) yeah, it's pretty slow. not just this session.
[15:27:12] <zulipbot> (Carsten Bormann) No, the integration seems to be really slow
Sometimes losing messages as well
[15:27:38] <zulipbot> (Kenneth Calvert) What is the advantage of doing this @ L3 vs L7?  I think this paper has looked at doing something similar @ L7: E. W. Zegura, M. H. Ammar, Zongming Fei and S. Bhattacharjee, "Application-layer anycasting: a server selection architecture and use in a replicated Web service," in IEEE/ACM Transactions on Networking, vol. 8, no. 4, pp. 455-466, Aug. 2000, doi: 10.1109/90.865074.
[15:30:24] <zulipbot> (Kenneth Calvert) Yes, it seems like the tail of the RCT may be of more interest?
[15:31:30] <zulipbot> (Colin Perkins) The IFIP Networking paper is presumably https://doi.org/10.23919/IFIPNetworking55013.2022.9829778
[15:32:23] <zulipbot> (Dirk Trossen) Thanks @Colin for posting the link!
[15:32:24] <zulipbot> (Colin Perkins) It's behing the IEEE paywall for me, but maybe there's a direct link
[15:33:02] <zulipbot> (Dirk Trossen) well, it used to for authors which was the link I accidentally posted on the list ;-)
[15:43:00] <zulipbot> (Dirk Trossen) @Kenneth sorry, only now realised your question at the bottom of the reference. Yes, the paper outlines the CDF to show the lowering of the variance through CArDS compared against the other mechanisms.
[15:43:53] <zulipbot> (Dirk Trossen) @Kenneth also note that CArDS only requires signaling the units used at deployment time, no regular metric reporting is necessary (such as suggested in the reference you sent)
[15:46:11] <zulipbot> (David Oran) Small point: Sidecars also act as a security perimeter, which is needed whenever you need to map between namespaces. So, unless yiou go full bore ICN, you still need something to do the mapping and isolation, and frankly sidecars seem to not actually have much overhead compared to other approaches.
[15:48:48] <zulipbot> (Jianfei(Jeffrey) HE) Could someone move the local camera to the presenter?
[15:48:48] <zulipbot> (Kenneth Calvert) @DirkT - got it, thanks.
[15:49:16] <zulipbot> (Jianfei(Jeffrey) HE) thanks!
[16:00:15] <zulipbot> (Kentaro Ebisawa) Tushar, what was the target HW of the Taurus data plane ? FPGA or Switch ASIC (e.g. Tofino) ?
[16:04:03] <zulipbot> (Colin Perkins) Have to run to another meeting - great talks!
[16:04:42] <zulipbot> (David Oran) further into the packet - and encrypted to boot!