The Grid

The Large Hadron Collider is finally turning on.


A quick step backwards: the LHC is a particle accelerator, the largest of its kind, underwritten by all the wild money in science, a ringed tunnel some 27 kilometers long, deep underground, crossing the French-Swiss border at four points. It's been over twenty years in the making and has garnered the support of 10,000 scientists in 85 countries behind its unimaginable modus operandi: to recreate the environment of our universe as it was less than a millionth of a second after the Big Bang, and hence to reveal, among other things, the fundamental nature of matter. By all accounts a significant accomplishment: that something as massive and as diplomatically enlightened as this machine could be made in such a fiercely nationalistic era, that the technology even exists, and that the sheer logistical nightmare of its operations could be overcome. And, while the imminent revelations of the LHC will undoubtedly chew up much of my scientific ruminations in the next few years, it's just these, the logistical operations, that I'm currently interested in.


This is because (unbeknownst to many) the LHC project has a second, more pragmatic, tentacle. It's called the Grid.

What is the Grid?

Some experts are calling it a "parallel internet." Although this is, in many ways, a reasonable moniker, the Grid is primarily the solution to one of the LHC's most important problems, which is the outrageous density of data it will begin to emit the second it goes live. Like, 15 Petabytes (15 million Gigabytes) of data annually, the analysis of which will ultimately require some 100,000 CPUs of processing power (NUMBERS!), which thousands of scientists around the world need to access and analyze in order to make a lick of sense of it. Rather than be stored on site at the CERN in Switzerland (the site of the LHC), this data needs to be distributed globally, parsed, narrowed down, and parceled out to the 7,000 physicists who need it.

How will it work?

Hence the Grid: a system of dedicated 10 gigabit per second fiber-optic cables connecting the Large Hadron Collider's crazy monumental magnetic detectors directly to the CERN computing center (or centre, if you will), then outwards throughout the world in a three-tiered system. The raw data is tossed into tape storage at CERN, then transmitted on these same fiber-optic cables to 11 "Tier One" research facilities, who are responsible for reprocessing the raw data and redistributing it.

Next down the line are the 150 "Tier Two" centers, mostly universities, which are located all around the world. The data arrives here via standard Internet protocols (i.e. using the regular ol' Internet, albeit in the guise of general purpose research networks, such as the U.S. Department of Energy's Energy Sciences Network), and is then disseminated to all the physicists for their invaluably real-live human analysis.

Here is a useful schematic for understanding how the Grid works, if you want to get more technical, i.e. see diagrams.

In any case, we're talking about 55,000 servers already installed, with another 145,000 on the way in the next two years. Remember when everyone was freaking out about the Googleplex? That's nothing. This fiber-optic network is 10,000 times faster than the fastest existing broadband. My friend Scott, who told me about the Grid, was like, "get ready for holographic video!" It's huge. I could throw around confusing approximations like, "it would take 25 days to transfer the nearly 400,000 movies on IMDB," but suffice to say it's a massive upgrade from the kinds of Internet speeds we're used to.

It seems strangely appropriate, strangely telling, that the CERN would implement this system. After all, the research facility was fundamental in implementing the Internet protocols that would bring about this first wave, that would enable me to sit here at my kitchen table and interface blindly with a nebulous and globally-distributed network of information, an absurdity in itself. I see it as inevitable that the Grid, or a system like it, is going to mold our communications, our media, our daily lives, in ways we can't possibly imagine or predict.

In his writings, the computer scientist -- and fabulist, although aren't they all, the good ones -- Vernor Vinge, no uncertain proponent of the ever-developing Technological Singularity theory, noted that "every time our ability to access information and to communicate it to others is improved, in some sense we have achieved an increase over natural intelligence." What he meant was that the end of the human era (which he argued would occur "[not before ] 2005 or after 2030") would come with a whimper, not a bang -- "even the largest avalanches are triggered by small things," he added.

I don't imagine that the Grid will go all Skynet on us, but if the history of the Internet tells us anything, it's that we can't predict, nor can we place enough expectations, on the exponential nature of its evolution. Besides, Vinge wasn't spooking us when he wrote, in his 1993 essay The Coming Technological Singularity: How to Survive in the PostâHuman Era, that "even the egalitarian view of an Internet that wakes up along with all mankind can be viewed as a nightmare." It is a nightmare, not least because it is strangely probable, but also because the Grid is so inextricably linked to the Large Hadron Collider, this fountainhead of certain scientific revolution, the two projects so potent with possibility, sinister and otherwise. Fellow science fiction heads will recognize this kind of setup from so many novels. The audacity of man is unbreakable.

And what if all the LHC reveals to us is that matter is only information broken down into infinitesimally small parts? We would already have begun to recreate it, a new universe slowly subsuming the last, only to awaken, unsolicited, in order to ask its own, similar questions about its place in the universe. It makes me feel crazy to think how profoundly the future refuses to remain at bay.

A prediction: even while the Large Hadron collider offers a final, unquestionable answer about the fundamental nature of the Universe, it's the Grid that will change the world, slipping in like a legislative footnote and blooming, guileless, the final nail in the coffin of the twentieth century.


More like this

The positive and sometimes unexpected impact of particle physics is well documented, from physicists inventing the World Wide Web to engineering the technology underlying life-saving magnetic resonance imaging (MRI) devices. But sometimes the raw power of huge experiments and scientific ambition…
William Gibson, first in his novel Burning Chrome and then later in the seminal Neuromancer, both coined and defined "cyberspace" as "a consensual hallucination experienced daily by billions of legitimate operators." His novels predate the universal adoption of  the World Wide Web as a…
For your reading and collection development pleasure. It's been so long since I last did one of these listings, I actually have another one coming up in a day or so. Good Faith Collaboration: The Culture of Wikipedia by Joseph Michael Reagle Jr. (ISBN-13: 978-0262014472) Wikipedia, the online…
The stakes are high, and win or lose, worth playing for. -Fred Hoyle Yesterday was a record-setting day for humanity. Over at CERN, the Large Hadron Collider achieved a new record for the most energy ever created in a laboratory collision in the history of humanity: 7,000,000,000,000 electron Volts…

Thank you. Incredible photo of The Grid, as well. Thank G for supercomputing.