- NANOG - lists.nanog.org

Java BGP stack
by Paul Jimenez 06 Oct '99

06 Oct '99

For your edification, researchers and toymakers, I'd like to point you at http://www.place.org/~pj/software, which has a pointer to a BGP stack written in java. Of course source is included. Have fun. --pj

1 0

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

Multicast is broken as an idea, period. Now, why won't we just forget about it and spend our life doing something useful instead? --vadim Now working for Alcatel :) Oui. > can a broken protocol design, msdp, be configred correctly? > > randy

5 5

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

ddivinia(a)broadcast.com (Darin Divinia) wrote: >Caching doesn't work for live content. Why? (And what is "live content"? You mean interactive? Or "live" TV coverage?) --vadim

7 9

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

Jared Mauch <jared(a)puck.nether.net> wrote: > The problem is that everyone thinks that multicasting is > more complicated than it really is. The problem is, everyone thinks that million lemmings can't be wrong. Multicasting cannot be made scalable. It is as simple as that. One can play with multicast routing protocols as much as one wishes - in pilots and small networks. It only takes a question - "ok, how are we going to handle 100000 multicast channels?" to realize that L3 multicasting is not going anywhere as a real-world production service. Yes, there are better multicasting schemes (like EXPRESS from Stanford, or even more scalable multicasting I described some time ago in TRAP documents). Still, they are not enough. > People try and overthink a few too many things, and this I > think is one. Nah. It is a clear and present case of not thinking hard enough. > Worse yet, it distracts from deployment of the real solution - cacheing. > Multicasting is faster than disk. This is a rather strange statement. I worked on a product (which was shipped) which delivered something like 20Gb/s of streaming video content from disk drives. RAID can be very fast :) > I'm not sure how caching > is the solution. Distributed content is also good. Ah, distributed content :) Yet another kludge requiring lots of maintenance and "special" relatioships between content providers and ISPs. --vadim

3 2

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

Sorry for misattribution and confused date... (silly me, i should be checking references before posting them and not relying on memory alone). Thanks, Vijay! --vadim From: Vijay Gill <wrath(a)cs.umbc.edu> On Tue, 5 Oct 1999, Vadim Antonov wrote: > ...and on '67 vintage theoretical work by Kleinrock and Klein who showed > that routing can be done in scalable manner. Aggregation is nothing new. Kleinrock and Kamoun. Most of the basis for the above are available in: Kleinrock, L. and F. Kamoun, "Hierarchical Routing for Large Networks, Performance Evaluation and Optimization", Computer Networks, Vol. 1, No. 3, pp. 155-174, January 1977 (A copy of this paper that Dr. Kleinrock was kind enough to autograph for me is one of my more prized possessions) And another other classic: Kamoun, F. and L. Kleinrock, "Stochastic Performance Evaluation of Hierarchical Routing for Large Networks", Computer Networks, Vol. 3, pp. 337-353, November 1979. /vijay

1 0

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

Sam Thomas <sthomas(a)lart.net> wrote: > I am not foolish enough to argue with a million lemmings. One need only > convince the one in front to change direction. You'll be amazed at the incompetence of lemmings in front :( > That a more scalable solution has not yet been developed is not evidence > that one does not exist. Of course, not. But this is someting in research domain - if and when a magic solution is found, then it'll make sense to discuss it. Doing engineering in hope of some future magic is, well, unwise. And there _are_ problems which cannot be solved. > Classful routing and address assignement wasn't > scalable, either, but we have (hopefully) come out of that era with some > cleverness and reliance on good design. ...and on '67 vintage theoretical work by Kleinrock and Klein who showed that routing can be done in scalable manner. Aggregation is nothing new. >> Nah. It is a clear and present case of not thinking hard enough. > Indeed, and you're willing to dismiss multicast without second thought? No, i'm willing to dismiss it after years of playing with it and thinking about ways to fix it. > As time passes, I can assure you that the line between the two will blur, > whether because of multicast, caching, or greed is yet to be seen. Greed wins. > * I am a multicast newbie, and largely illiterate in current implementation, > so don't laugh at my suggestions publicly, please. :-) Uncommon honesty :) --vadim

2 1

Re: Real Media and M-Bone feeds
by Vadim Antonov 06 Oct '99

06 Oct '99

Fredrik Nyman <fredrik(a)swip.net> wrote: >> Multicasting cannot be made scalable. It is as simple as that. > Think posivite! And see comment below. I love that. Do you seriously think that positive thinking can change the fundamental problem of having traffic-dependent global routing information? We're having lots of problems with traffic-independent routing information already, thank you. > How are you going to handle 100000 realvideo streams? If all those 100000 > sources are in the US and all recievers are in Europe, there would be no > difference in the amount of bandwidth required across the ocean with > multicast compared to realvideo. Easily. It is called cacheing. Multicasting provides exactly the same bandwidth savings as cacheing with cache retention time set to zero. Now - how are you going to handle the same users who want to watch newsreel from the beginning, so as not to miss the important news they caught the end of? We already have cable companies transmiting the same content with different delays many times over. That's one of a hell of efficient technology. > And I'd definitely prefer to use multicast when I bring all the worlds > radio and TV-channels to all the worlds users. Yep. Sure. The only problem that _nobody_ needs or wants simultaneous transmissions. Just because it is the common practice due to medium limitations, it does not mean it is desireable or has any market potential. We ain't need no second television. The old one is cheap and widely deployed, thank you. Note that i left the issue of oligocasting (multicasting to few parties, such as useful in conference calls) out of the picture. Oligocasting at L3 has even worse scalability problems, and the only realistic approach at this time seems to be using MCUs - like the telephony people do for decades. > Given time and enough resources multicast will rock. I would prefer backbones to work steadily, without any rocking motions. > Would you have > thought 10 years ago that there'd ever be something called world wide web, > and a zillion users on the Internet? 10 years ago i definitely was a believer - starting the first ISP in a country spanning 11 time zones :) > How could you possibly scale the > Internet using IGSs and the occational AGS? Can't. 12000s and M-40s aren't much better, either. That's why i invented the terabit routers :) --vadim

3 2

Re: Real Media and M-Bone feeds
by Vadim Antonov 05 Oct '99

05 Oct '99

Sam Thomas <sthomas(a)lart.net> wrote: > What would be the point of this? > ...If you suggest a hierachical cache (i.e. main > cache takes a feed, replicates to n distributed caches closer to the > receivers) you have just re-invented multicast, and not very well. ;-) >Those who do not understand Unix are condemned to reinvent it, poorly. > -- Henry Spencer Those who do not understand routing are condemned to reinvent it, poorly. --vadim

1 0

Re: Real Media and M-Bone feeds
by Vadim Antonov 05 Oct '99

05 Oct '99

Charles Sprickman <spork(a)inch.com> wrote: > And wouldn't a cache box with multiple people watching "live content" > basically bring us right back to multicasting? One stream to the cache > operator, multiple streams out to the clients... Multicasting. There's a subtle but very important difference: such "multicasting" does not involve injection of customer-originated routing information into the global routing system. Which means that scalability is not an issue. > There's at least one commercial service that does this with RV and > NetShow. Yep. --vadim

1 0

Re: Real Media and M-Bone feeds
by Vadim Antonov 05 Oct '99

05 Oct '99

From: Randy Bush <randy(a)psg.com> >> Multicast is broken as an idea, period. Now, why won't we >> just forget about it and spend our life doing something >> useful instead? > because, although it is getting less expensive quickly, transport costs > money. multicast promises to reduce that cost near sources. I strongly suspect that the combined time of engineering staff spent on deployment and development of mulitcasting won't be paid off by the supposed transport cost savings any time soon. Worse yet, it distracts from deployment of the real solution - cacheing. > this is a better fantasy than the qos smokers who think it will effectively > get more bits into the pipe. ...or at least more buds :) --vadim

3 3