New 'Grid' to replace World Wide Web

[CaptainChewbacca]

Times Online

Coming soon: superfast internet

THE internet could soon be made obsolete. The scientists who pioneered it have now built a lightning-fast replacement capable of downloading entire feature films within seconds.

At speeds about 10,000 times faster than a typical broadband connection, “the grid” will be able to send the entire Rolling Stones back catalogue from Britain to Japan in less than two seconds.

The latest spin-off from Cern, the particle physics centre that created the web, the grid could also provide the kind of power needed to transmit holographic images; allow instant online gaming with hundreds of thousands of players; and offer high-definition video telephony for the price of a local call.

David Britton, professor of physics at Glasgow University and a leading figure in the grid project, believes grid technologies could “revolutionise” society. “With this kind of computing power, future generations will have the ability to collaborate and communicate in ways older people like me cannot even imagine,” he said.

The power of the grid will become apparent this summer after what scientists at Cern have termed their “red button” day - the switching-on of the Large Hadron Collider (LHC), the new particle accelerator built to probe the origin of the universe. The grid will be activated at the same time to capture the data it generates.

Cern, based near Geneva, started the grid computing project seven years ago when researchers realised the LHC would generate annual data equivalent to 56m CDs - enough to make a stack 40 miles high.

This meant that scientists at Cern - where Sir Tim Berners-Lee invented the web in 1989 - would no longer be able to use his creation for fear of causing a global collapse.

This is because the internet has evolved by linking together a hotchpotch of cables and routing equipment, much of which was originally designed for telephone calls and therefore lacks the capacity for high-speed data transmission.

By contrast, the grid has been built with dedicated fibre optic cables and modern routing centres, meaning there are no outdated components to slow the deluge of data. The 55,000 servers already installed are expected to rise to 200,000 within the next two years.

Professor Tony Doyle, technical director of the grid project, said: “We need so much processing power, there would even be an issue about getting enough electricity to run the computers if they were all at Cern. The only answer was a new network powerful enough to send the data instantly to research centres in other countries.”

That network, in effect a parallel internet, is now built, using fibre optic cables that run from Cern to 11 centres in the United States, Canada, the Far East, Europe and around the world.

One terminates at the Rutherford Appleton laboratory at Harwell in Oxfordshire.

From each centre, further connections radiate out to a host of other research institutions using existing high-speed academic networks.

It means Britain alone has 8,000 servers on the grid system – so that any student or academic will theoretically be able to hook up to the grid rather than the internet from this autumn.

Ian Bird, project leader for Cern’s high-speed computing project, said grid technology could make the internet so fast that people would stop using desktop computers to store information and entrust it all to the internet.

“It will lead to what’s known as cloud computing, where people keep all their information online and access it from anywhere,” he said.

Computers on the grid can also transmit data at lightning speed. This will allow researchers facing heavy processing tasks to call on the assistance of thousands of other computers around the world. The aim is to eliminate the dreaded “frozen screen” experienced by internet users who ask their machine to handle too much information.

The real goal of the grid is, however, to work with the LHC in tracking down nature’s most elusive particle, the Higgs boson. Predicted in theory but never yet found, the Higgs is supposed to be what gives matter mass.

The LHC has been designed to hunt out this particle - but even at optimum performance it will generate only a few thousand of the particles a year. Analysing the mountain of data will be such a large task that it will keep even the grid’s huge capacity busy for years to come.

Although the grid itself is unlikely to be directly available to domestic internet users, many telecoms providers and businesses are already introducing its pioneering technologies. One of the most potent is so-called dynamic switching, which creates a dedicated channel for internet users trying to download large volumes of data such as films. In theory this would give a standard desktop computer the ability to download a movie in five seconds rather than the current three hours or so.

Additionally, the grid is being made available to dozens of other academic researchers including astronomers and molecular biologists.

It has already been used to help design new drugs against malaria, the mosquito-borne disease that kills 1m people worldwide each year. Researchers used the grid to analyse 140m compounds - a task that would have taken a standard internet-linked PC 420 years.

“Projects like the grid will bring huge changes in business and society as well as science,” Doyle said.

“Holographic video conferencing is not that far away. Online gaming could evolve to include many thousands of people, and social networking could become the main way we communicate.

“The history of the internet shows you cannot predict its real impacts but we know they will be huge.”

Sounds REALLY interesting, if it pans out.

[PeZook]

Isn't it almost exactly how the Internet got started out?

The only thing I see problematic is that you basically need to build a new infrastructure for that, whereas the Internet is accessible to anybody with a phone line (heh, this is the same basic idea, it just has better tubes)

Of ocurse, it's not like we didn't build massive infrastructural projects before, but it always took a couple of years.

Or decades

[cosmicalstorm]

What speeds exactly are they talking about?

Here in Sweden you can get a 100/100 megabit/s connection for about $50 a month, in some places there's even the possibility of a 1 Gbit/s connection for those who are willing to pay the extra buck.

[CaptainChewbacca]

What speeds exactly are they talking about?

Here in Sweden you can get a 100/100 megabit/s connection for about $50 a month, in some places there's even the possibility of a 1 Gbit/s connection for those who are willing to pay the extra buck.

The article says 10,000 x broadband. If you were getting 100 meg/sec, you will now be getting 1000 gigabytes (1 terrabyte) per second.

[cosmicalstorm]

Megabit, not megabyte.
They probably meant 10.000 x 1 megabit, i.e. 10 Gbit/s, thats not too farfetched.
Though I have a hard time seeing how this will be implemented everywhere as long as most American ISP's continues to operates under the "what can we get away with?" policy.

[Guardsman Bass]

Assuming that it can be expanded, what they ought to do is make an all-out effort to get it installed in universities. They tend to be good places to start out with that kind of stuff.

[Eleas]

The article says 10,000 x broadband. If you were getting 100 meg/sec, you will now be getting 1000 gigabytes (1 terrabyte) per second.

The problem is, what with the sloppy labeling of even the suckiest ADSL or ISDN lines as "broadband", this description is so vague as to be meaningless. "The typical broadband connection"? Oh, really? Typical in what sense? There are vast differences from region to region.

[EnsGabe]

I'm quite amused by this article. While I don't deny that massive increases in bandwidth is an impressive technical feat, billing it as a 'whole new internet' is a laughably simplistic view. Was the move from using 28.8kbps modems to high speed cable internet access a 'whole new internet'? While the capabilities that the move afford can change the way you use the internet, it's a far cry from being a new internet. Adding new high capacity links between parts of the network is hardly a revolutionary idea.

In addition, it looks to me like the writer of the article was conflating 'Grid Computing' with new networking capability, which is understandable since the former requires the latter.

[Rye]

Good, maybe now fucking driver sites will go beyond the 5k/s upload rates. This reminds me of some kooky google conspiracy flash animation I saw ages ago that predicted the internet would be upgraded into a monitored system called "the grid."

[NoXion]

THE internet could soon be made obsolete. The scientists who pioneered it have now built a lightning-fast replacement capable of downloading entire feature films within seconds.

At speeds about 10,000 times faster than a typical broadband connection, “the grid” will be able to send the entire Rolling Stones back catalogue from Britain to Japan in less than two seconds.

the RIAA and MPAA are NOT going to be happy about this...

[Rahvin]

I'm quite amused by this article. While I don't deny that massive increases in bandwidth is an impressive technical feat, billing it as a 'whole new internet' is a laughably simplistic view. Was the move from using 28.8kbps modems to high speed cable internet access a 'whole new internet'? While the capabilities that the move afford can change the way you use the internet, it's a far cry from being a new internet. Adding new high capacity links between parts of the network is hardly a revolutionary idea.

In addition, it looks to me like the writer of the article was conflating 'Grid Computing' with new networking capability, which is understandable since the former requires the latter.

It's a completely different network, not just a connection speed increase. Instead of piggybacking on existing telecom lines like the current internet does, they're talking about a physically separate network utilizing dedicated fiber channels and, from what it sounds like, possibly different routing methods.

They aren't just talking about giving you fiber access at your home - they're talking about a completely different network designed from the ground up. Your analogy to increased modem speeds is irrelevant, as this article is not discussing increased speeds on the client side, but rather a completely redesigned network that functions independantly of the current internet.

[Patrick Degan]

Lovely, but the technical detail uppermost on the minds of many users will be how fast they'll be able to download their porn.

[cosmicalstorm]

There was a story about Japan planning to build a network after these principles a couple of months ago.

Japan plans research for new network to replace Internet

The Associated Press
Tuesday, August 28, 2007

TOKYO: Japan will start research on new network technology to replace the Internet to tackle growing quality and security problems, a government official said Tuesday.

U.S. and European researchers have already started research to rebuild the underlying architecture of the Internet, a move that could mean replacing networking equipment and rewriting software on computers.

Yoshihiro Onishi, assistant director at the Japanese communications ministry, said Japan feels it's crucial to follow suit to stay competitive. Post-Internet network technology is expected to become imperative by 2020, he said.

"The Internet is reaching its limit," he said. "We feel this research for the technology is definitely needed."

The ministry is now working to set up a research organization by fall next year. It will request a budget for the fiscal year starting April 1, 2008, although the amount of money has yet to be decided.

Cooperating on the research with other nations, including the U.S., will be an option, Onishi said.

Among the concerns researchers may address is the need to set up an ecological network that consumes low levels of energy.

With the arrival of hackers and spammers, and the proliferation of laptops and other mobile devices, scientists are starting to believe a totally new network is needed.

The new network may run parallel with the Internet or eventually replace it, or parts of the research could go into a major overhaul of the existing architecture.

In the U.S., such efforts are still in their early stages and aren't expected to bear fruit for another 10 or 15 years. Congress has yet to come up with funding for the projects.

http://www.iht.com/articles/ap/2007/08/ ... ternet.php

[Mr Bean]

Grid Details Below
The Theoretical top speed of "The Grid" was clocked at 17 Tb, yes I said Tera-bits. Even if acutal real world speed is only 1% of that it's a huge fing increase over what we have today.

I'm quite amused by this article. While I don't deny that massive increases in bandwidth is an impressive technical feat, billing it as a 'whole new internet' is a laughably simplistic view. Was the move from using 28.8kbps modems to high speed cable internet access a 'whole new internet'? While the capabilities that the move afford can change the way you use the internet, it's a far cry from being a new internet. Adding new high capacity links between parts of the network is hardly a revolutionary idea.

Do you remember 28.8kbps days? Do you remember waiting five-ten seconds between loading just plain text pages? Try the Internet Way back machine, how about old Yahoo? 1997
I still remember that taking about four whole seconds to load on a ISDN 64k modem and being so impress on how much faster it was over a 28.bps modem I had used before.

The Internet has fundamentally changed several times, what it was about, what it had on it and how easy it was to find information.

3Gbps is the limit, both because it's a nice number and because at that speed you can easily stream full 1600x1200 video with enough left over to handle VOIP, Instant Messaging and a dozen browser windows with ease.

Once we hit that limit it will be possible to once again re-design the Internet since things that are impossible in the past(IE high quality video streaming) will become easy. Could something like You-tube have existed in 1996? No the speed's were not yet up to it yet. Now it is, and soon it will be much closer to DVD or even Blue-ray quality Streaming video instead of 320x320 video. We are close to stepping into a high-def world on the Internet, and once we hit that point the only other places to go are VR or Holographic.

So will this fundamentally change the internet? Beat your ass it will, but not as much as the step that will come after.

[phongn]

It's a completely different network, not just a connection speed increase. Instead of piggybacking on existing telecom lines like the current internet does, they're talking about a physically separate network utilizing dedicated fiber channels and, from what it sounds like, possibly different routing methods.

By and large, the Internet does not piggyback on telecom lines but runs on its own dedicated fibre networks. Only the "last mile" tends to be on the PSTN, and even then only for DSL and dial-up.

The Theoretical top speed of "The Grid" was clocked at 17 Tb, yes I said Tera-bits. Even if actual real world speed is only 1% of that it's a huge fing increase over what we have today.

Speeds like that will probably be limited to backbone lines. End-to-end speed will be a tiny fraction of that (probably significantly less than 1 Gb/sec).

The Internet has fundamentally changed several times, what it was about, what it had on it and how easy it was to find information.

There hasn't been a fundamental change to the Internet in years. The world wide web has been in a state of continuous change and occasional revolution (i.e. "Web 2.0") but that's different.

3Gbps is the limit, both because it's a nice number and because at that speed you can easily stream full 1600x1200 video with enough left over to handle VOIP, Instant Messaging and a dozen browser windows with ease.

It just sounds like you pulled that number out of your ass. 1080p video can be streamed with much less than 100 Mb/sec of bandwidth.

We are close to stepping into a high-def world on the Internet, and once we hit that point the only other places to go are VR or Holographic.

Bandwidth isn't the problem for VR or holography.

[Mr Bean]

It just sounds like you pulled that number out of your ass. 1080p video can be streamed with much less than 100 Mb/sec of bandwidth.

I'll got that off Tom's hardware so I'll admit the information is unsourced at best. Personal experience has told me standard 10/100 Ethernet you can't do 1600x1200 streaming without some issues.

[Sriad]

Order of magnitude check:

Without compression algorithms, 1600x1200 would require
1600x1200 (pixels) x 24(bits per pixel for 16 million colors) x24(frames per second)=1,105,920,000 bits per second.

You can pull this down with good compression, but TRUE streaming HD will require that much bandwidth.

[phongn]

I'll got that off Tom's hardware so I'll admit the information is unsourced at best. Personal experience has told me standard 10/100 Ethernet you can't do 1600x1200 streaming without some issues.

Given good NICs and switches, you can easily stream HD video at 100 Mb/sec.

You can pull this down with good compression, but TRUE streaming HD will require that much bandwidth.

No, it won't. First of all, there's nothing requiring that "true" HD (whatever "true" means) be raw video. It would be absolutely absurd to think that way: Blu-Ray maxes out at 40 Mb/sec, for example. Are you going to claim that it isn't "true" HD?

[Uraniun235]

This "grid" sounds a hell of a lot like Internet2.

[LadyTevar]

Mr Bean, I recall 300baud days, when 'plain text' was all that was available and 1200baud was impressively fast. You know ... 1988? (PS, your link didn't go anywhere but a log-in page)

17TB is mind-boggling to me.

[phongn]

This "grid" sounds a hell of a lot like Internet2.

There are not a few high-speed, limited-access networks popping up for academic and scientific use. Internet2 is pretty much one of the older ones and the most well-known.

[Sriad]

You can pull this down with good compression, but TRUE streaming HD will require that much bandwidth.

No, it won't. First of all, there's nothing requiring that "true" HD (whatever "true" means) be raw video. It would be absolutely absurd to think that way: Blu-Ray maxes out at 40 Mb/sec, for example. Are you going to claim that it isn't "true" HD?[/quote]

For some reason I was thinking of "streaming" in terms of HD Camera->destination computer over the Grid instead of HD Camera->YouHDTube (and extremely intense CODEC)->Surfer.

...But my responsive research shows that HDV cameras still only record at 1440x1080 with a data output of 100Mbit/s.

[Sriad]

Too much editing: the broken quote in my above post is of phongn compelling me to educate myself on HD technologies.

[phongn]

...But my responsive research shows that HDV cameras still only record at 1440x1080 with a data output of 100Mbit/s.

Well, consumer-level equipment is limited to that. Professional equipment can have significantly higher bitrates. HDCAM SR HQ, for example, has a bitrate of 880 Mb/sec.

[Sea Skimmer]

Holy shit man, I’ll be able to get a thousand viruses and eighty different spy ware programs downloaded and installed in less time then it takes to close a browser window… this will truly be a glorious step forward.