Article In The Guardian On Internet2

Sam Halter writes: "The Guardian carries this story about the future of the Internet and the expermental Super high-speed academic networks that are being built in Europe and the U.S."

Comments

This article isn't about Internet 2

[Fjord]

The title Article In The Guardian On Internet2 is erroneous. The article is actually about Geant, "the new pan-European network serving more than 3,000 of the continent's academic and research institutions". Basically, Europes version of I2.

Internet2

[iGawyn]

I'm currently a student at the University of Maryland we have Internet2 connectivity. Stolen from OIT's network throughput page is:

A 75Mbps connection to Qwest Communications for commodity internet traffic.

An Gigabit Ethernet connection to MAX (Mid Atlantic Crossroads), a consortium of local research institutions, through which we have high speed connectivity to those institutions as well as the NSF vBNS network and the Internet2 Abilene Network.

An ATM connection to UMATS, the intercampus network of the University System of Maryland for connectivity with other USM schools.

Internet2 gives me downloads very close to the theoretical max of the 10megabit connection to my room, which is great for being an ultra-low-ping bastard in games. With the gigabit connection, the ping times to basically any location on Internet2 is the same as if it was on your LAN.

To answer the IPv4 vs. IPv6 question, it uses the same IPv4 that the rest of the world uses, it just appears to be more infrastructure to speed up things.

Gawyn

Misleading motivations for I2

[hyrdra]

People are getting it all wrong. I2 isn't the "next" or "new" Internet and it wasn't created for brand new applications or new "mindsets" for doing things because it's so blistering fast. It was created because schools can't afford commercial pipes. It's less expensive to connect schools together directly than connect them to a national, commercial provider at these high speeds.

I2 is primarily fast because it isn't used all that much. You don't have thousands of AOL dialups clogging up the network, @Home/Time Warner boxen downloading pirated movies, or the psychic friends network using it to do their VoIP. That all eats bandwidth. Instead, you have the occasional geek downloading a slackware distribution, or browsing the Computer Science department of another university. If suddenly all the schools would allow traffic over their commercial pipes to access their I2 routers, I'm sure the network would slow down now seeing it's accessible to the public - along with all the abuses and bandwidth eating applications.

I guess the best analogy to this would be comparing it to an underground tunnel between schools only for academic use, compared to a giant highway for public use. The underground tunnel doesn't nearly have the capacity of the massive highway, but is much faster. So just because something is fast doesn't mean its on the edge of technology or is, in fact, anything special.

I have used I2 and it is quite fast, but what can you get on it? The latest well hyperlinked personal page of a student in a nearby school? This makes it loose much of the reason why the real Internet is so popular -- it's a space where you can find anything. But I2 defeats this purpose by limiting what the network can connect to, and thus its usefulness. It may be useful at testing new applications, like an HDTV stream, but since you're not doing this on a public network to begin with it's applications are limited to your own, highly restrictive network. You can't say you've done something new when all you've done is create an exclusive network that doesn't address the real problems of networks anyway - like last mile access and exponentional bandwidth increases. IMO, I2 is a way for schools to have a fast link to each other without paying the huge costs associated with a 1 GB link to the national backbone. That's all it is, and that's all it probably will be.