- Test - lists.nanog.org

Re: Creating exchanges
by Barry Raveendran Greene 14 Oct '96

14 Oct '96

Hello All, This topic of Guam as an IX has come up many times in the past. Frankly, if it made since, it would have already happened. The #1 factor for an IX location in the Asia Pacific is the business case. Geographic topology does play a factor, but 1/2 circuit lease line prices are the major determinant if people will come to an IX. Guam's problem is that people have to pay the full circuit cost to get to Guam. These circuit costs are roughly 10X the cost of circuits in the US. In addition, pricing of the circuits are NOT distance sensitive. So for most of the region, the full circuit costs to Guam will be more expensive than a circuit to the US. One engineer/colleague working for a large US Internet backbone did work through the business case of placing a major hub in Guam for his Internet, FR, and X.25 traffic. He had the advantage of knowing what 'military' resources he can draw from (i.e. Typhoon hardened FM facilities) case he use to work for the Air Force in the region. The results was that the business case did not make since. Using Japan, HK, Singapore, and Australia as redundant hubs made better since. And that is what happened when they upgraded their backbone. Robert Mathews-ICICX wrote: >> From an infrastructure perspective, though, Japan looks hard to beat. >> The pipes to Singapore/Jakarta/Australia, Guam/Hawaii/L.A.-ish, and >> somewhere in Oregon (?) all meet there. Ignoring regulations, tariff >> issues, etc., of course. Japan and Singapore does look good because of the cable interconnects and the satellite foot prints. AIH/Abone (Japan - http://www.aih.net) and STIX/SIB (Singapore - http://www.stix.net) are two successes of international IXs - through both are commercial IXs - not neutral. Yet, each of them are working. Personally, after working this issue for 3 1/2 years, international neutral IXs will not work in Asia Pacific. National IXs are fantastic and work wonders. Commercial IXs are there to compete with the big US backbones in the region. Barry -- Barry Raveendran Greene | || || | Senior Consulting Engineering | || || | Singapore | |||| |||| | tel: +65 738-5535 ext 235 | ..:||||||:..:||||||:.. | e-mail: bgreene(a)cisco.com | c i s c o S y s t e m s |

3 2

Re: Internet: Getting more money ...??
by Dana Hudes 14 Oct '96

14 Oct '96

Point may have been obscured. You add more facts (thanks!) the question is what the $10 million represented -- did it include the money IBM spent but did not charge the NSF? As for the NSS's being still in place but disconnected, I thought the lease was up a year ago. The NSS's certainly are not cost effective. One of the problems that they faced in the versions subsequent to those you had at ANS was running out of backplane bandwidth for WAN-LAN (and LAN-LAN) transfers. Another of course was their proprietary nature. The price tag doesn't surprise me, Rs/6ks are expensive. But the sampling feature would be helpful in the SYN attack situation. Dana

2 1

Digest Possible??
by Andrew Smith 14 Oct '96

14 Oct '96

This question has probably been asked many times over, but I haven't been able to find it or an answer. Is there, and if there isn't, is getting it setup possible, a nanog-digest mailing list. --------------------------------------------------------------------------- Andrew Smith ** awsmith(a)neosoft.com ** Network Engineer ** 1-888-NEOSOFT ** "Opportunities multiply as they are seized" - Sun Tzu ** ** http://www.neosoft.com/neosoft/staff/andrew ** ---------------------------------------------------------------------------

1 0

Re: BBN outage
by rmg＠ranma.com 14 Oct '96

14 Oct '96

From: Nathan Stratton <nathan(a)netrail.net> > > So BBN says they are now going to install a generator, my question is why > was a generator not installed? They saved 20 - 100K on a generator, but I > bet they will lose more then that when users leave. Forsythe Hall, where the BBN pop is located, has a standby generator, abet not hooked up, to provide power when the Stanford CoGenerator facility goes off, like it did. The bldg UPS has plenty of power to keep the facilities running until the standby generator can be hooked up (BBN shares physical facilities with Stanford's IBM mainframe and the Stamford DMS-100 telephone switch). When I used to work there, I saw that generator all the time. It collected a lot of dust. I was told that it was tested out from time to time :) I wonder why it either 1) Did not work, or 2) Was not available ... Across the street, at the hospital, there is another generator set up on a trailer. When the power was out, that one was not running. I would have thought that could have been disconnected and brought over to Forsythe Hall to provide power for when the UPS's died off. Telco wise, BBN does have a nice dirverse system. Their MCI T-3 is on a microwave that goes to Palo Alto, where it drops off to MCI's bay area fiber ring between San Jose to San Francisco. Some of their backbone T-3's come into a Pac Bell OC-48, which has an entirely seperate protection path into Stanford itself (Stanford alltogether has two Pac Bell OC-48's, with a third one planned!). Another fiber path belongs to MFS for more dirversity. So on that level, they have their act together. r.

1 0

Re: Internet II is coming... (fwd)
by Oded Comay 14 Oct '96

14 Oct '96

My apologies for sending it to all of you. I am not sure who is interested in the subject. Traffic prioritization as implemented by routers is built around an interface output queue. In the case where a router is connecting a fast LAN to a relatively slow line, the queue might get built on the slow line interface (in the outgoing direction) but not the other way around. In the even more special case (but probably the common one) where the traffic is asymmetric so incoming traffic is significantly larger than outgoing one, no queue will be created, and the prioritization scheme will not be operated at all. Hope this helps, Oded. >----- Forwarded message from Michael Ciavarella ----- > >> ---------- Forwarded message ---------- >> Date: Wed, 9 Oct 96 07:04:08 UTC >> From: Hank Nussbacher <hank(a)ibm.net.il> >> To: Sean Doran <smd(a)cesium.clock.org>, "Dorian R. Kim" <dorian(a)cic.net> >> Cc: nanog(a)merit.edu >> Subject: Re: Internet II is coming... >> >> ..... >> >> The reason for this is that within the university, access and use is unlimited >> and uncontrolled. The student and the professor have equal access from their >> workstation. That is why the Israeli university consortium has come up with >> a different solution. It is called chokepoint. A unix system that acts >> as a firewall/gateway. If the total access speed to the Internet is >> T1 then at the chokepoint one can define that port80 can use a maximum of >> 700kb. And one can define that telnet is guaranteed 30kb. And that >> 10.2.1.1 is guaranteed 128kb no matter what. This way, faculty server and >> faculty workstations can be given priority over student access. In addition, >> no one faculty member can "hog" the system. Facultycan even pay to the >> chokepoint to improve their service over others. An entire set of rules >> based on protocol and IP address can be set up and implemented. They have >> been running this way for the past 6 months. >> > >How does this differ from the queuing and prioritisation features on, >say, a cisco? Obviously router load is reduced by moving this type of >"service" to a dedicated box, but unless you're using the UNIX box as >your router to the net, how does it exert this type of control over your >bandwidth? Does it drop incoming packets which exceed a (set) threshold >even though they've already come over your link, or does it sit >at your provider's end, in which case how do you control your outgoing >traffic? > > >mike.. (blah. mornings.) > >----- End of forwarded message from Michael Ciavarella ----- >

1 0

Re: Intra/Inter - was Inet-II
by Paul Ferguson 14 Oct '96

14 Oct '96

I forgot to include my original smiley. ;-) - paul At 11:48 AM 10/14/96 -0400, Curtis Villamizar wrote: >> Another suggestion? Lacking one, one might suggest that the complexity >> of the 'mesh' needs to be simplified. >> >> - paul > > >Paul, > >It's that sloppy multiprovider model. Let's go back to one NSFNET >core. :-) > >Curtis > >ps- For the *extremely* humor impaired -- I'm just joking. I'm not >seriously suggesting that. I am suggesting to Paul through (friendly) >sarcasm that we have to live with routing complexity. >

1 0

Re: Intra/Inter - was Inet-II
by Havard.Eidnes＠runit.sintef.no 14 Oct '96

14 Oct '96

> ftp://www.isi.edu/div7/ra/Publications/bgp_osc.ps.gz This one should be ftp://ftp.isi.edu/ra/Publications/bgp_osc.ps.gz - Havard

1 0

Re: generators, etc....
by Steve Norton 14 Oct '96

14 Oct '96

At 11:56 AM 10/14/96 -0400, Curtis Villamizar wrote: >If we hold BBN to the same standards that the telco industry uses to >come up with 99.999%, BBN was up but "a few customers" experienced a >localized outage. Discussion of reliability standards has gotten a little vague. To throw some facts into the confusion: I once managed to beat our Telco reps hard enough for them to provide actual numbers. This was after a 4 hour outage on an OC-3 ring that "could self heal in 50 milliseconds." Ameritech provided the usual mushroom treatment, but MFS actually quoted their Operations procedure AP2402, which says: For a DS3 facility 0-50 miles: Availability on a 12 month period is 99.997% Error free seconds > 99.992% over 24 hour period (7 seconds/day) Bit error rate < 2.5*10E-11 "Performance tests shall be conducted at DS-3 level for 72 hours." Apparently this is copied from AT&T PUB 62411. Cisco gives me the mushroom treatment, but assuming they shoot for something like 99.992% uptime, a simple DS3 Internet connection could have no better than 99.98% reliability. Does anyone even come close to this number? With all of the ATM deployed recently, I cant imagine any ISP even comes close. Steve Norton 312-496-4601 VP Operations 312-496-4499 (fax) InterAccess Co steve(a)noc.interaccess.com

2 1

Re: Intra/Inter - was Inet-II
by Paul Ferguson 14 Oct '96

14 Oct '96

Herr Manning, At 11:15 AM 10/11/96 -0700, bmanning(a)ISI.EDU wrote: >> >> You mean that routing protocols need to take account of policies? ;-) >> >> Or do you mean that BGP isn't good enough? >> >> Michael Dillon - ISP & Internet Consulting > >Yes... (silly :) >Yes... (serious) > >BGP has a number of serious flaws. The more complex the routing mesh, the >more they will become visable. And if what I see is correct, IDRP, the >BGP heir, has the same sets of problems. > Another suggestion? Lacking one, one might suggest that the complexity of the 'mesh' needs to be simplified. - paul

4 5

Re: Internet II is coming...
by salo＠msc.edu 14 Oct '96

14 Oct '96

> Subject: Re: Internet II is coming... > Date: Wed, 09 Oct 1996 17:20:17 -0400 > From: Fletcher E Kittredge <fkittred(a)biddeford.com> > > On Wed, 9 Oct 1996 12:13:03 -0700 (PDT) Michael Dillon wrote: > > My ISP has a 10Mbps fibre ATM circuit to BCNet in Vancouver that was > > installed in Aplril 1995. From there it connects to CA*Net which has a T3 > > into MCI Seattle as well as links to the East where more T3's head south > > to MCI. The T3's were T1's until Sept 1995 at which point there was a > > small improvement in speed at times. > > So a 10Mbps fiber ATM circuit translates, after the cell tax and > packet stredding, to 6Mbps on a real link. Is this not slow enough to > be a bottleneck? And, an OC-12c ATM circuit can provide nearly 540 Mbps of TCP data, after SONET, ATM, IP and TCP overheads. Perhaps, I missed your point... -tjs

2 1