September 1997 - Test - lists.nanog.org

Re: Traffic Engineering (fwd)
by Sean M. Doran 19 Dec '97

19 Dec '97

Eric Germann <ekgermann(a)cctec.com> writes: > At 04:31 PM 9/18/97 -0400, Sean M. Doran wrote: > > > >Perhaps you could explain to me how you can find the > >shortest path between A and B using ping times, traceroute > >hop counts, and AS_PATHS observed at C, assuming that > >traffic between A and B is not exchanged through C? > > > > You're not trying to find it between A and B. A connects to B, but B has > every intention of redirecting to C1, C2, or C3, etc. The originally proposed idea was that search engines would present a list of hits sorted by the proximity to the client that made the query. What you are doing is rehashing the IBM scheme. Unfortunately the IBM scheme requires * an initial connection to a server * each server must know the location of its replicated data * each server must be able to determine the proximity of its replicated data to the client in reasonable time and without placing too much additional load on the server, the client, or the replicating sites Point (1) means that for short transactions, preserving locality is a net loss. If it is more work to redirect than to serve directly, there is no point in not serving directly. Point (2) means that copies of the data that may be nearer the client may not be known about. This is also a scaling problem. Point (3) is a difficult engineering challenge, not least because determining proximity (assuming proximity is a function of bandwidth, delay, and infrastructure) without inducing traffic load is hard. What is really needed is a scheme such that * the client learns the identity of the datum desired, and the ultimate, authoritative source of the datum, through user interaction (typing in a URL) or through a search database * the client or its proxy asks increasingly distant infrastructure if replicas are available; Van Jacobson suggests this is a good application for multicast and I'm inclined to agree somewhat * the client or its proxy retreives the found replicated datum if a replica exists and is found within a reasonable time or scope of multicast search or the client or its proxy turns to the ultimate source for a copy * should the client or its proxy hear a query for this same datum, it will offer it up for retrieval to other clients that is, everything in your client or proxy's local cache may be served up to nearby clients Note that the squid caching system emulates this scheme to a large degree. The difficulty is now reduced to constructing a spanning tree for all multicast queries and determining whether something found by the client or its proxy to be topologically nearby is a good choice for serving up the data. These problems seem much more tractable than the ones the IBM scheme needs to solve, and a great deal of theoretical work has gone into this sort of model already. > Anyone know of some reasonably available methods for > measuring end to end "performance" which are almost > universally implemented? No. See the CAIDA or IPPM efforts for why not. Sean. P.S.: Curtis Villamizar had another interesting approach which involved pushing content far afield to machines with the same transport-layer (IP) addresses, relying upon closest-exit routing to connect one to the topologically-closest replication machine. Unfortunately, while this could be really cool for NSPs to offload stuff towards peering points (public or private), it also has some poor scaling properties and is uncomfortably reliant upon the stability of routing. If he's done any more thinking about the idea, I'd love to hear about it though.

4 4

Re: Traffic Engineering
by James R Grinter 17 Dec '97

17 Dec '97

freedman(a)netaxs.com (Avi Freedman) writes: > smd(a)clock.org (Sean Doran) writes: > > P.S.: Curtis Villamizar had another interesting approach > > which involved pushing content far afield to > > machines with the same transport-layer (IP) > > addresses, relying upon closest-exit routing to > > connect one to the topologically-closest replication > > machine. Unfortunately, while this could be really (this is something we've done a little of at work: we do have it in production use, and it's not so effective when there's asymmetric routing - which we have quite a bit for some European networks, eg two peerings with Ebone. Duplicating IP addresses definitely has its place, eg internal to a network for directing a bunch of RASs to a local version of a 'service' address.) > And Alec Peterson (now of Erols) has figured out an even > arguably slicker way to do it. This is a problem I've been musing over lately. The best way that I can see is to return suitable DNS answers to people. This avoids the route instability problems that Sean comments on, above: Client gets TCP session established. Routes change, their inbound route now reaches a different system which clearly doesn't know about the connection and *boom* goes the TCP session. Routes return to their previous situation but by this time it's too late. (Even worse is when the client connects *during* the route instability.) but with DNS you've still got to return a suitable answer. (You could route to your DNS server, but that doesn't address the asymmetry problems.) If the bulk of your data is from you to the client (web, euuw), then it's ok because you are in a position to decide what the best route back across your network is and pick the nearest/most appropriate server. Requests usually come from the client's nearest recursing/resolving DNS server and so that's just a SMOP to decide the best way back. Where you get really screwed up is having to support large numbers of IP addresses - good ol' HTTP/1.0 requests (it's that damn web, again). Using (n x number of "sites") isn't practical. Maybe the real operational issue is working out when we stop having to assign all these zillions of IP addresses for web servers. The marketting people don't want to make the change whilst their competitors aren't. James.

4 4

ATM (was Re: too many routes)
by Sean M. Doran 09 Oct '97

09 Oct '97

Richard Irving <rirving(a)onecall.net> writes: > Ok. I will bite, although I hate to open my mouth, as my shoe always > seems to bee-line for it.. ;} Hehe. > I don't think so.... how about the ability to mix > voice, MPEG, and IP on the same pipe ? Um, I do this now with IP. Admittedly with parallel circuits (virtual or hard) I could send such traffic down different pipes to partition congestion effects, however to do this right I'd really want to use MPLS/tag switching anyway. When I make the decision to use MPLS/tag switching I also have to consider that there is decent queuing available in modern IP routers (that will effectively become hybrid IP routers and MPLS/tag switches) and that I can neatly partition the traffic without using actual or virtual circuits. > Or, how about that with ABR my delay across the ATM > fabric is reduced when I have more bandwidth open. (POTS > is low on utilization, during this "theoretical moment > in time") A couple milliseconds and a few extra Mbs can > count ;) You want bounded delay on some traffic profiles that approach having hard real time requirements. (Anything that has actual hard real time requirements has no business being on a statistically multiplexed network, no matter what the multiplexing fabric is). This can be implemented in routers now, with or without MPLS/tag switching, although having the latter likely makes configuration and maintenance easier. ABR is another attempt to do statistical multiplexing over a substrate that is not well geared to anything other than TDM. It interacts poorly with any protocol that is developed to run over a statistically-multiplexed network (e.g. TCP) and there are little demons with respect to the way RMs are handled that can lead to nasty cases where you really don't get the bandwidth you ought to. The problem again is that mixing TCP and other statmux-smart protocols with ABR introduces two parallel control loops that have no means of communication other than the interaction of varying traffic load, varying delay, and lost data. This often leads to the correct design of additive increase/multiplicative decrease traffic rate response to available bandwidth leading to a stair-step or vacillation as more bandwidth becomes available to each control loop, and a rather serious backing off at the higher level when available bandwidth is decreased even fairly gently. Delay across any fabric of any decent size is largely determined by the speed of light. Therefore, unless ABR is deliberately inducing queueing delays, there is no way your delay can be decreased when you send lots of traffic unless the ATM people have found a way to accelerate photons given enough pressure in the queues. > Oh, 2 things come to mind, my variability throughout an ATM cloud is > greatly reduced versus a routing cloud, a cell requires WAY less time to > cross a switches backplane, versus a packet through a router. And > seriuosly less time to determine where to send it... Um you need to be going very slowly and have huge packets for the passage through a backplane to have any meaning compared to the centisecon propagation delays observed on long distance paths. I know of no modern router that delays packets for anything approaching a handful of microseconds on fast interfaces except in the presence of outbound queues being congested, where if you're running TCP you really want to induce delay rather anyway, so that the transmitter will slow down. > Ok. So, maybe Cisco's Flow Switching approaches VBR having a bad hair > day. (and tuned for SERIOUS tolerance, CDVT=10,000), but certainly not > traditional routing. The analogy between VBR and flow switching confuses me. Could you explain this a bit? Actually, maybe you could explain the rest of the message too, because I think we have a disconnect in terms of vocabulary. :( Sean.

25 37

Re: *scream* Cannot contact AT&T WorldNet NOC
by Sean Donelan 01 Oct '97

01 Oct '97

If you don't directly connect or peer with them, have you tried going through your upstream provider to get a trouble ticket referral for their NOC? AT&T has been very good about providing the secret code-word and telephone number to their direct inter-connects and peers in the past. >Does *anyone* have a contact number for someone that does not keep >asking me if I'm running a Windows machine? Every number I have for >them, connects me to some moron that has no clue when I say "It >appears that someone from your network is attempting to flood our >network". Asking to be connected to Security gets me an e-mail >address and asking for a supervisor get me "He's not avainable" Whois >on att.net get me a voice mailbox saying that they are closed on the >weekends. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

4 4

IE4 goes live; Imminent Death of the Net predicted
by Jay R. Ashworth 30 Sep '97

30 Sep '97

Advance warning: -----Forwarded message from "Charles Oppermann" <chuckop(a)MICROSOFT.com>----- From: "Charles (Chuck) Oppermann" <chuckop(a)MICROSOFT.com> Subject: IE4 goes live @ 7pm PT Date: Tue, 30 Sep 1997 14:10:43 -0700 Dear friends, Internet Explorer version 4.0 goes live on http://microsoft.com in between 6pm and 7pm PT today, September 30! This is a product that I've worked on and I'm very proud of. In case you're interested, my contribution to the product was the accessibility options dialog box, various UI and HTML features, and exposing the Dynamic HTML Object Model through Microsoft Active Accessibility. [ snippage ] The following Internet Service Providers (ISPs) are also offering IE4 from their sites: * Concentric * AT&T * EarthLink * Erol's * GTE * MCI * MindSpring * NETCOM * Prodigy * Sprint * SPRYNET It's a big download - 22mb approximately for the full installation - so, you might want to just order the CD-ROM from Microsoft. Information on doing that is at the above listed web sites. -----End of forwarded message----- Y'all have fun now, y'hear? >:-) Cheers -- jra -- Jay R. Ashworth jra(a)baylink.com Member of the Technical Staff Unsolicited Commercial Emailers Sued The Suncoast Freenet "People propose, science studies, technology Tampa Bay, Florida conforms." -- Dr. Don Norman +1 813 790 7592

1 0

Re: DS3 PAIX-MaeWest
by jc＠crl.com 30 Sep '97

30 Sep '97

Andrew - long time no talk! Talk to Paul Berggren. He might be able to set something up... -Jim

1 0

DS3 PAIX-MaeWest
by andrew khoo 30 Sep '97

30 Sep '97

i am in the process of provisioning a DS3 from PAIX to Santa Clara that i will be using about 15Mbps of at max. does anybody want/need to share a chunk of this DS3? ============================================================== andrew khoo - rebel leader gsm: +61-500-555-466 Ourworld Global Network, Australia work: +61-2-9223-2777 sushi-eating good guy - wookie avi pager: +61-2-9962-4414 speaks in home-grown tongue - ren usa: +1-415-531-2754 ============================================================== -- We Bite (tm) --

1 0

Re: *scream* Cannot contact AT&T WorldNet NOC
by Sean Donelan 29 Sep '97

29 Sep '97

>> Alternatives: >> - Call the FBI Computer Crime Squad (202) 324-9164, which >> will probally be very unsatisfying unless you can document >> a large dollar loss > >Try Byron Thompson from NC-17. You may have much better luck Last time Sprint did this, I believe a front page story in the New York Times was very effective at getting their attention. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

2 1

MAE East outage?
by Alex Rubenstein 29 Sep '97

29 Sep '97

Hello MAE-EAST? thank you nitrous.digex.net Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State 165.117.1.17 4 2548 6934754 1496216 13625659 0 0 4w6d 165.117.1.35 4 2548 5907855 1526342 13625659 0 0 22:14:25 165.117.1.122 4 2548 1532453 3211484 13625556 0 0 4w6d 165.117.1.127 4 2548 60052 11744465 13625779 0 13 4w6d 192.41.177.2 4 1 49656 92380 0 0 0 00:50:52 Active 192.41.177.3 4 6478 63566 81892 0 0 0 00:50:52 Active 192.41.177.65 4 6205 99295 141169 0 0 0 00:50:52 Active 192.41.177.68 4 6113 165184 96560 0 0 0 00:50:52 Active 192.41.177.75 4 3951 50131 90229 0 0 0 00:50:52 Active 192.41.177.80 4 5000 50972 90919 0 0 0 00:50:51 Active 192.41.177.85 4 3491 71324 90136 0 0 0 00:50:51 Active 192.41.177.86 4 4565 51673 90927 0 0 0 00:50:51 Active 192.41.177.90 4 5388 50453 93455 0 0 0 00:50:51 Active 192.41.177.92 4 3914 49874 91278 0 0 0 00:50:51 Active 192.41.177.104 4 2041 93901 94947 0 0 0 00:50:51 Active 192.41.177.110 4 2685 58082 90924 0 0 0 00:50:51 Active 192.41.177.112 4 5378 52425 98304 0 0 0 00:50:50 Active 192.41.177.118 4 6196 49874 90710 0 0 0 00:50:50 Active 192.41.177.119 4 3967 51872 90732 0 0 0 00:50:50 Active 192.41.177.120 4 286 79607 90348 0 0 0 00:50:50 Active 192.41.177.121 4 2914 22465 27131 0 0 0 00:50:50 Active 192.41.177.125 4 297 66117 91477 0 0 0 00:50:50 Active 192.41.177.141 4 1673 101843 108025 0 0 0 00:50:50 Active 192.41.177.145 4 4200 153354 93300 0 0 0 00:50:50 Active 192.41.177.147 4 3847 65196 91088 0 0 0 00:50:49 Active 192.41.177.148 4 6172 49864 91375 0 0 0 00:50:49 Active 192.41.177.150 4 4136 147781 91894 0 0 0 00:50:49 Active 192.41.177.160 4 3407 147540 92424 0 0 0 00:50:49 Active 192.41.177.166 4 2885 49092 91508 0 0 0 00:50:49 Active 192.41.177.169 4 2885 49808 90347 0 0 0 00:50:49 Active 192.41.177.181 4 3561 580412 93460 0 0 0 00:50:49 Active 192.41.177.182 4 5683 66842 91811 0 0 0 00:50:49 Active 192.41.177.195 4 5390 49862 97760 0 0 0 00:50:48 Active 192.41.177.210 4 2551 71309 91828 0 0 0 00:50:48 Active 192.41.177.226 4 1746 50533 91515 0 0 0 00:50:48 Active 192.41.177.228 4 4006 50285 92109 0 0 0 00:50:48 Active 192.41.177.240 4 1800 471186 92147 0 0 0 00:50:48 Active 192.41.177.241 4 1239 829269 82275 0 0 0 00:50:47 Active 192.41.177.242 4 1239 814619 82034 0 0 0 00:50:47 Active 192.41.177.245 4 174 20408 16644 0 0 0 00:50:47 Active 192.41.177.248 4 701 1259066 93690 0 0 0 00:50:46 Active 192.41.177.251 4 293 54973 90461 0 0 0 00:50:46 Active 198.32.186.2 4 1 212647 102411 0 0 0 00:50:46 Active 206.205.242.7 4 3706 57639 99946 13625779 0 0 1w3d

1 0

Netcom?
by Alex Rubenstein 29 Sep '97

29 Sep '97

Does anyone have a magic number to NetCom NOC or abuse center? Thanks!

8 7