- ripe-list - mailman.ripe.net

Re: Very fast IP, no ATM...
by Peter Lothberg 09 Dec '96

09 Dec '96

> Could anybody help me a bit with SONET/SDH technology: > what is the present state of - say STM4c vs STM4 etc.- > standardization. I think you are mixing Sonet and SDH terminology here, both SDH and Sonet are vell standardized upto 622. Speed SDH Sonet ----------------------------------------- 155M STM-1 OC3c 622M STM-4 OC12c > At Stuttgart we run some 20km dark fibre driven by FORE`s > ATM/STM4c boards (NEC SX4 32 - Cray T3E 512) - would > we buy e.g. NORTEL SDH equipment if this would be - > say 100km - and what about public European SDH? I can't answer, as you mix names. But, for the two speeds mentioned above, both Sonet and SDH transport equipment could be configured to carry either of. The tricky part is how to configure protection, and that's the same for both Sonet and SDH. > How is vBNS going to upgrade to 622Mbit/s ? (Not my department) but I think they use ATM switches. > What does the the below "Cisco's Native IP over SDH/VC4" > exactly say? Read the RFC on PPP over Sonet, but here is how it's stuffed into a container, and note I don't do PPP on the link we talked about, just cisco HDLC. |<--------------------------------------------------270 Bytes-------->| |<--------------9 Bytes------------>|<1B>|<---------260 Bytes-------->| -- +---------------------------------------------------------------------+ | A1| A1| A1| A2| A2| A2| J0| Z0| Z0||J1|| | |-----------------------------------||--|| + SOH | B1| | | E1| | | F1| | ||B3|| | +-----------------------------------||--|| HDLC byte stuffed + | D1| | | D2| | | D3| | ||C2|| inserted here | -- +-----------------------------------||--|| + -- +-----------------------------------||--|| Synchronous Payload + Point| H1| H1| H1| H2| H2| H2| H3| H3| H3||G1|| Envelope (SPE) | -- +-----------------------------------||--|| + -- +-----------------------------------||--|| + | B2| B2| B2| K1| | | K2| | ||F2|| ----------------------------||--|| + | D4| | | D5| | | D6| | ||H4|| | +-----------------------------------||--|| + LOH | D7| | | D8| | | D9| | ||Z3|| | +-----------------------------------||--|| + |D10| | |D11| | |D12| | ||Z4|| | +-----------------------------------||--|| + | S1| Z1| Z1| Z2| Z2| M1| E2| | ||Z5|| | -- +---------------------------------------------------------------------+ --Peter

1 0

SIRCE Developments (long)
by Daniel Karrenberg 09 Dec '96

09 Dec '96

Dear contributors, dear colleagues, the final state of commitments on December 2nd was just short of ECU 40k coming from 44 sources. Thank you all! Organisation name Person commit (ECU) pi.se AB Pontus Ekman 1000 ACONET Wilfried Woeber 1000 RedIRIS-CSIC Victor Castelo 1000 Academic Computer Centre, CYFRONET-KRAKO Karol Franczak 500 IPGlobal, Informatica e Telecomunicacoes Pedro Ramalho Carlos 500 Teleport C&S GmbH. Christoph Maerk 500 BT Public Internet Service Nigel Roy Titley 1000 Tele Greenland Michael Schultz 1000 SpaceNet GmbH Sebastian v. Bomhard 750 Unisource Business Networks Italy Mauro Magrassi 1000 Sontheimer Datentechnik GmbH Bernd Sontheimer 500 ECRC GmbH Dave Morton 1000 Tele2/SWIPnet Jorgen Ericsson 1000 Nacamar Klaus Landefeld 1000 SIBS - Sociedade Interbancaria de Servic Jose Eduardo Pina Mi 1000 LITNET Daiva Tamulioniene 500 INRIA Annie Renard 500 EUnet Communications Services BV Per Gregers Bilse 1500 HogiaNet Kenneth Hybner 1000 AT&T Internet Services Tankut Turhan, Henk 1000 CESNET Pavel Vachek 500 OTEnet S.A. Theodore Thanopoulos 1000 HEAnet John Hayden 1000 U-NET Limited Stuart Muckley 500 Telia AB J=F6rgen I Larsson 1000 GTN Andreas Baess 1000 INS GmbH Andreas Frackowiak 1000 NASK - Research and Academic Network in Krzysztof Silicki 500 Xlink Michael Rotert 1000 Entreprise des Postes et Telecommunicati Thierry Coutelier 500 HPT - Croatian Post and Telecomm, Teleco Ivan Sedinic 1000 Unisource Business Networks (Schweiz) AG Philip Bridge 1000 Clinet Heikki Suonsivu 1000 InComA Ltd. Gregory Dmitriev 1000 Internet ProLink Mickey Coggins 500 MAZ Internet Services Rainer Lillge 1000 IFS Daniel David 1000 Unisource Business Networks NL bv Rob Reitsma 1000 Global One Sweden Eric Malmstrom 1000 LF.net GmbH i.G. Kurt Jaeger 500 NLnet Services bv Aris Doelman 1000 Transpac France/Global One Bernard Malaval 2000 I.NET S.p.A Marco Negri 1000 Telecom Finland Pasi-Mikko Rautanen 1000 ----- Total: 39250 ===== This represents just 10% of the current contributors and about a quarter of the amount we considered necessary to justify the project as proposed in ripe-150. From this we have to conclude that at this time there is not enough interest in the service among our current constituency to justify such an ambitious pilot service carried out at the RIPE NCC. I have therefore withdrawn the NCC's proposal. Obviously there is some disappointment because we would have liked to do the job. On the other hand I am quite happy that we did not embark on this project with very short preparation only to find out then that the support was less than expected. We will let you know when TERENA has decided about the award of the SIRCE project. You can then consider supporting the particular effort TERENA will select. I'd also like to thank the 51 people who responded to the questionnaire. Here are the results: n % 51 100 Total 47 92 STI likes that NCC asks feedback this way 46 90 ST9 We are satisfied with current NCC services 44 86 STG questionanaire easy to understand 35 69 STA RIPE should start security coord WG 35 69 ST4 have looked at ripe-150 (proposal) 26 51 ST6 not enough time to consider 22 43 ST5 should have been discussed at RIPE meeting 11 22 STB do not contribute because too vague 11 22 ST2 SIRCE needed in a couple of years 9 18 STF TERENA involvement makes it too complicated 9 18 STE do not contribute because too expensive 7 14 STD do not contribute because we do not need it 6 12 STJ do not contribute until selctin process complete 4 8 ST1 SIRCE not needed at this time 3 6 STC do not contribute because NCC not the right place 3 6 ST8 NCC should not start new services 3 6 ST7 NCC should not do incident coordination 0 0 STH questionnaire took too much time 0 0 ST3 SIRCE not useful at all This further supports that the SIRCE service is currently not perceived as very urgent by many ISPs but that it may become more important in the future. A clear lesson I personally have learned from this is that the consensus building process within RIPE and the NCC contributors committee is a delicate one which takes time and careful consideration by all concerned. It is a good process which has served us very well so far. I will take great care to respect it better than I did in this particular matter. The NCC team and I will now do our best to increase the number of you who are saisfied with our current services from a good 90% to as near to 100% as we can make it. Regards D. Karrenberg

1 0

Re: Very fast IP, no ATM...
by huitema＠bellcore.com 05 Dec '96

05 Dec '96

Paul, You asked me to repeat that: if not the telephone system itself- would Christian prepare an analysis saying "a Internet of `appropriate` capacity with RSVP and IPng would allow for concurrent AUDIO (as Jon said) of phone quality for everybody ..." The only problem I have with this phrase is the mention of RSVP. I don't think there is actually any major problem for providing a best effort service dimensioned to fit the needs of 600Mio internauts (or 6 billions, for that matter). I am convinced that some policing will be needed to make sure that usage is proportionate to investments (and subscription rates). I am however definitely not convinced that RSVP is what we need there. RSVP is about giving special priviledges to a few flows; we need to provision sufficient resources so that the average flow has adequate capacity, which is a different problem. -- Christian Huitema

2 1

corrected UK figures for October
by Mirjam Kuehne 04 Dec '96

04 Dec '96

Dear all, I forgot to send the updated figures for the October count for the UK. The decrease of hosts that we saw in October was probably due to problems reaching the other TLD UK domains while the hostcount was running. The count was then repeated on Tue, 12 Nov 96 and showed the follwoing results: CY SOA COUNTED DUPL REAL CHANGE ============================================================ "wrong" count for October: uk 31468 697703 98605 599098 -49118 updated count for October: uk 32567 796010 101282 694728 +46512 Please note that this also changes the CHANGE column for the UK for November: uk 34363 815905 106068 709837 +15109 I copied the updated version of the UK hostcount to ftp://ftp.ripe.net/ripe/hostcount/uk.{output,error}-Oct.gz Kind Regards, Mirjam Kuehne RIPE NCC

1 0

Re: Very fast IP, no ATM...
by Paul Bryant 04 Dec '96

04 Dec '96

At 15:36 03/12/96 +0000, you wrote: > > > >Interestingly BT claim that their network will never overload as they charge > >for local calls - ie ration by cost. The corollary is that if the network > >becomes overloaded they raise the cost. > > >a voice network nowadays is pretty easily overengineered anyhow Well - some countries. > >afterall, people don;t spend more than 8 hours a day on the phohne, >and if one person is speaking, the other is silient, so you only have >to support 12th of the worlds population of voice traffic - if yo udo >silence suppression I think you may be wrong. On a national network (modern) the network is digital and once you are past the codec you don't know whether the call is voice or ISDN data. If you start to do silence suppression on ISDN I suspect that things may go wrong. Internationally you are right - not only do you have silence suppression but also voice compression to 16K. That is why on international ISDN you have to dial 000 rather than 00 for an international call to ensure digital 64K end to end. Interestingly this doubles the cost. Taking into account speech may be compressed and silence suppressed double the cost looks cheap (but don't tell BT). I suspect that 28K modems do not work well internationally and I suspect that the new 56K modems (do we call them modems?) do not work at all. Am I right? > Paul Bryant - Rutherford Appleton Laboratory, Chilton, Didcot, OX11 0QX, UK Mail P.Bryant(a)rl.ac.uk Tel.+44 1235 445 267 Fax. +44 1235 446 626

1 0

Re: Very fast IP, no ATM...
by Paul Christ 04 Dec '96

04 Dec '96

Peter, congratulations again. - Could anybody help me a bit with SONET/SDH technology: what is the present state of - say STM4c vs STM4 etc.- standardization. At Stuttgart we run some 20km dark fibre driven by FORE`s ATM/STM4c boards (NEC SX4 32 - Cray T3E 512) - would we buy e.g. NORTEL SDH equipment if this would be - say 100km - and what about public European SDH? How is vBNS going to upgrade to 622Mbit/s ? What does the the below "Cisco's Native IP over SDH/VC4" exactly say? Paul ---- p.s.: ... if not the telephone system itself- would Christian prepare an analysis saying "a 600Mio participants Internet of `appropriate` capacity with RSVP and IPng would allow for concurrent AUDIO (as Jon said) of phone quality for everybody ..." :-) ------------------------------------------- On Dec 2, 2:28pm, Peter Lothberg wrote: > Subject: Re: Very fast IP, no ATM... > > p.s.: have you any measures of IP-level throughput over this link? > > Almost 2 moths of operation, 4h downtime and we have been > running upto 120Mbit/s user traffic. > > (Traffic generated by traffic generators does not count, but > it's possible to fill it up to theoretical max.) > > It's just a releif to see this first link that have enough > buffer memory to deal with the network burstiness and allow > TCP to do it's job.. > > --Peter > > > >On Monday 23 September 1996 00:15Z.. > > > > > >Was the worlds first transatlantic 155Mbit native IP service brought into > > >operation by Sprint/USA and Tele2/Sweden. > > > > > >The circiut wich is part of Sprintlink and ICMnet runs between the > > >NY-Nap in Pennsauken, NJ, USA and Tele2 in Sweden and uses cisco > > >packet-over-sonet/sdh technology. (Native IP over SDH/VC4) >-- End of excerpt from Peter Lothberg

3 2

Subscription
by Anfilova Blanka 04 Dec '96

04 Dec '96

I wish to subscribe for the RIPE mailing list. Thank you. Blanka Anfilova (Mrs.) nic-hdl: BA85-RIPE NCC-Reg.ID: cz.cznet

1 0

Hostcount for November 1996
by Mirjam Kuehne 03 Dec '96

03 Dec '96

Dear all, Here is the hostcount for November 1996. The table shows an increase since last month of over 82,000 (or 2.5%) to a total of almost 3,450,000 hosts in the DNS in Europe. The UK shows the highest increase for November with more than 110,000 hosts more than in October (Remember the amount of hosts in the UK dropped significantly last month, now it is back to normal :) In some countries we see quite a significant decrease this month, also in countries were the hostcount is done remotely. We couldn't find any explanation for this besides general connectivity problems and servers not responding. Please note that FI consistently loses hosts (hidden behind firewalls?). BA is the first time included in the count. The hostcount can also be found under ftp://ftp.ripe.net/ripe/hostcount. The graphical version of the hostcount can be found at ftp://ftp.ripe.net/ripe/hostcount/RIPE-Hostcount-{Mo,Qu}.eps The translation of the country codes can be found at ftp://ftp.ripe.net/iso3166-countrycodes All this data can also be found at http://www.ripe.net/statistics/ I would like to thank all people who help me to create the hostcount every month by counting hosts locally: at Gerhard Winkler <winkler(a)cc.univie.ac.at> be Herman Moons <Herman.Moons(a)CS.kuleuven.ac.be> ch Marcel Schneider <hostmaster(a)nic.ch> cy Panagiotis Tzounakis <pj(a)ics.forth.gr> cz Jiri Orsag <Jiri.Orsag(a)eunet.cz> de Peter Koch <pk(a)TechFak.Uni-Bielefeld.DE> dk Robert Martin-Legene <robert(a)dknet.dk> es Jose Sanchez-Seco <jose.sanchez(a)rediris.es> fi Petri Ojala <ojala(a)eunet.fi> fr Annie Renard <Annie.Renard(a)inria.fr> gr Dimitris Zisiadis <zisiadis(a)forthnet.gr> hr Ivan Maric <Ivan.Maric(a)stribor.srce.hr> hu Gabor Kiss <kissg(a)sztaki.hu> ie Mike Norris <mnorris(a)dalkey.hea.ie> il Sason Levi <sason(a)NetVision.net.il> it Antonio_Blasco Bonito <bonito(a)nis.garr.it> no Havard Eidnes <Havard.Eidnes(a)runit.sintef.no> pl Irek Neska <irek(a)nask.pl> pt Armando Domingues <fccn(a)dns.pt> ru Andrey Dobrotsvetov <adbr(a)ripn.net> sk Ivan Lescak <Ivan.Lescak(a)Slovakia.EU.net> Lubos Elias <Lubos.Elias(a)uakom.sk> sm Antonio_Blasco Bonito <bonito(a)nis.garr.it> tr Yener Yigit <yener(a)knidos.cc.metu.edu.tr> uk Kevin Hoadley <K.Hoadley(a)nosc.ja.net> Please note, that the next hostcount (December 1996) will be published later than usual, approximately mid January. This is due to my holidays :) I hope this doesn't cause any inconveniences. If you have any questions, please don't hesitate to contact me. Kind regards, Mirjam Kuehne RIPE NCC RIPE DNS Hostcount Previous Count : Thu Oct 31 1996 This Count : Sat Nov 30 1996 CY SOA COUNTED DUPL REAL CHANGE ============================================================ al 5 78 0 78 - 3 am 9 202 5 197 + 4 at 3706 97376 11849 85527 + 4591 az 6 49 1 48 + 29 ba 7 30 0 30 + 30 be 3475 65523 2534 62989 + 3050 bg 126 3168 151 3017 + 38 by 14 251 10 241 + 41 ch 6586 133636 6982 126654 + 7379 cy 149 1168 18 1150 + 104 cz 1555 40797 1362 39435 + 420 de 23980 727211 39091 688120 +35557 dk 4394 106028 4126 101902 + 7301 dz 1 0 0 0 - 18 ee 275 5729 117 5612 - 2312 eg 97 1969 109 1860 + 124 es 3296 110509 1943 108566 + 7344 fi 2560 199894 3558 196336 -112855 fo 23 96 0 96 + 7 fr 5522 228077 8104 219973 - 7211 gb 2 40 0 40 0 ge 13 188 2 186 + 19 gr 522 12339 644 11695 - 3559 hr 424 4508 733 3775 - 260 hu 908 32152 1666 30486 + 2226 ie 1368 26064 484 25580 + 618 il 1853 39648 3213 36435 - 5554 is 298 11033 111 10922 + 458 it 7306 154899 13173 141726 -10488 li 88 272 77 195 + 11 lt 190 1809 112 1697 + 44 lu 289 4062 774 3288 - 76 lv 159 5719 296 5423 + 697 ma 28 528 65 463 + 27 md 10 85 4 81 + 61 mk 21 237 4 233 + 22 mt 68 558 88 470 + 67 nl 9189 249097 9190 239907 + 7618 no 5340 153728 6082 147646 + 6132 pl 1970 52106 2506 49600 + 1459 pt 850 22695 771 21924 + 1483 ro 412 5447 186 5261 + 494 ru 1977 42373 3068 39305 + 2979 se 10234 249645 14278 235367 + 9425 si 563 26262 1136 25126 +12762 sk 394 7580 383 7197 + 127 sm 12 720 328 392 + 28 su 568 18375 895 17480 - 199 tn 4 44 3 41 - 1 tr 727 14135 768 13367 + 738 ua 306 7961 1505 6456 + 461 uk 34363 815905 106068 709837 +110739 va 2 5 0 5 0 yu 187 1904 86 1818 + 371 ============================================================ 136431 3683914 248659 3435255 +82549

1 0

Re: Very fast IP, no ATM...
by Paul Bryant 03 Dec '96

03 Dec '96

At 08:48 03/12/96 +0000, you wrote: > > > >You are missing the point, to make voice work, you have to move bits > >between sender and receiver, and, it does not really matter how you do > >it. > > >More interesting is to se when you can no longer make a normal > >phone-call, becouse all the resources in the voice switch is consumed > >by people accessing the Internet with Modems or ISDN. > > >What will they do, improve the voice network (very expensive) or give > >people real IP services? Once they given people real IP services, > >well, voice is just an application. Interestingly BT claim that their network will never overload as they charge for local calls - ie ration by cost. The corollary is that if the network becomes overloaded they raise the cost. > > >absolutely - voice (and hi hi audio) work very nicely on our 155 Mbps >IP over ATM net - they could ONLY work BETTER if we tok the ATM >out....in fact, the utilisation on our IP on 34Mbps SMDS is far >better...and even there, i think we could sacrifice the SMDS if it >wasn't so stable... > > jon > > Paul Bryant - Rutherford Appleton Laboratory, Chilton, Didcot, OX11 0QX, UK Mail P.Bryant(a)rl.ac.uk Tel.+44 1235 445 267 Fax. +44 1235 446 626

2 1

Re: Very fast IP, no ATM...
by Peter Lothberg 03 Dec '96

03 Dec '96

> * We can run the world-wide telephone system say by IPv6 * > * > > over SDH? > * > What make you believe we could not ? * > * > Christian Huitema > > Hereby I formally declare > > > Christian THE Ultimate Internet Prophet > > > Don't miss to deposit his visionary statement in a safe place. > > > Artigst > > > Paul Christ :-) Paul, You are missing the point, to make voice work, you have to move bits between sender and receiver, and, it does not really matter how you do it. More interesting is to se when you can no longer make a normal phone-call, becouse all the resources in the voice switch is consumed by people accessing the Internet with Modems or ISDN. What will they do, improve the voice network (very expensive) or give people real IP services? Once they given people real IP services, well, voice is just an application. --Peter

2 1