The Information Factories Are Here

prostoalex writes, "Wired magazine has coined a new term for the massive data centers built in the Pacific Northwest by Google, Microsoft, and Yahoo! Cloudware is, ironically, a return of the centralized data and bandwidth power houses caused by the decentralized and distributed nature of the Internet. George Gilder thinks we're witnessing something monumental: 'According to Bell's law, every decade a new class of computer emerges from a hundredfold drop in the price of processing power. As we approach a billionth of a cent per byte of storage, and pennies per gigabit per second of bandwidth, what kind of machine labors to be born? How will we feed it? How will it be tamed? And how soon will it, in its inevitable turn, become a dinosaur?'"

Comments

Supply of fiber too low for a revolution?

[Salvance]

At some point massive data centers won't provide incremental benefits unless the massive increases in processing power are met with proportional decreases in bandwidth prices. Sure, bandwidth prices have dropped, but not by nearly the rate of price/teraflop processing has. Companies like Google recognize this, and are investing in their own fiber to compensate. But the telecommuncations companies are the ones that originally build these lines, and it's unfortunately in their best interest to keep the supply of spare bandwidth very low.

Why Gilder Is Telecosmically Wrong

[miller60]

Everything is getting cheaper but power, which for some data centers now costs more than hardware. Nicholas Carr explains why Gilder's assumptions are problematic:

"What Gilder calls 'petascale computing' is anything but free. The marginal cost of supplying a dose of processing power or a chunk of storage may be infinitesimal, but the fixed costs of petascale computing are very, very high. Led by web-computing giants like Google, Microsoft, Amazon, and Ask.com, companies are dumping billions of dollars of capital into constructing utility-class computing centers. And keeping those centers running requires, as Gilder himself notes, the "awesome consumption" of electricity"

As I noted in our commentary at Data Center Knowledge, the power issues with high-density blade server computing has been understood for years. Back in 2002, Liebert and APC and other equipment vendors were developing products that could address huge heat loads. They saw it coming, and sensed a market opportunity. So where were the chip makers? Even as cooling vendors prepared for the results of the huge power and heat loads, little was done to address their source.

The last page of TFA...

[d474]

This is pretty cool writing:

"The next wave of innovation will compress today's parallel solutions in an evolutionary convergence of electronics and optics: 3-D and even holographic memory cells; lasers inscribed on the tops of chips, replacing copper pins with streams of photons; and all-optical networks in which thousands of colors of light travel along a single fiber. As these advances find their way into an increasing variety of devices, the petascale computer will shrink from a dinosaur to a teleputer - the successor to today's handhelds - in your ear or in your signal path. It will access a variety of searchers and servers, enabling participation in metaverses beyond the ken of even Ray Kurzweil's prophetic imagination. Moreover, it will link to trillions of sensors around the globe, giving it a constant knowledge of the physical state of the world, from traffic conditions to the workings of your own biomachine."

Makes me want to read a William Gibson novel.

Re:The last page of TFA...

[radtea]

As these advances find their way into an increasing variety of devices, the petascale computer will shrink from a dinosaur to a teleputer - the successor to today's handhelds - in your ear or in your signal path.

Technological prognostications are almost always wrong in two directions.

1) The ability of current tech to scale up indefinitely is always eventually proven false. For six decades new aircraft designed increased their average crusing speed from about 100 mph in 1920 to 700 mph in the 70's. Then they stopped getting faster, and have been at just under 600 mph ever since. The jump to 700 mph in the '70's was with the introduction of the Concord, which also gave the average crusing speed a huge variance. No one knows what the "speed of sound" for computing will be, and as always it will be a matter of economics rather than pure technological possibility, but what we do know is that it is out there somewhere, and eventually we will hit it, although possibly not for some decades yet.

2) The uses to which tech is put, the directions and consequences of the speed and size improvements that do happen, are always almost completely wrong, as are the costs. As Asimov once said, the challenge is not to predict the automobile but the parking problem. Lots of people predicted some of the social consequences of the 'Net, but I don't think anyone predicted spam. These sorts of things may create limits that come into play earlier than other economic limits (and not incidently, create major opportunities for companies with solutions able to overcome them.)

As near as I can tell, the "parking problem" of the brave new world of ubiquitous interconnected computing, is the identity issue. Computers deal with proxies for everything. Unlike human beings, they are a realization of Plato's Cave, dealing only with the numerical shadows of reality. And one shadow can be made to look very much like another.

It doesn't matter

[sillybilly]

It doesn't matter if computing performance doubles if the software that runs on it decays in performance at an even greater rate. Back in 1988 MSDOS used to boot in less than 10 seconds after the BIOS POST. Who cares if you'll have software with features greater than your brain, with capacity to even guess your thoughts, wishes and desires, if it will just do what you want without mouseclicks of speech commands, who cares for all these features if it takes 5 years to boot up on a computer a gazillion times faster than today's computers, and its processing speed is uttering 3 words per decade while consuming 900 gigawatt of electric power? Case in point: Windows Vista.

Yes and no. Mostly you missed the point, sorry

[Moraelin]

Even if you take the meteor hypothesis as absolute truth, the fact is: other species survived. Not only mammals, but also lizards. Heck, even some species of dinosaurs survived. (Birds _are_ technically dinosaurs.)

We're not talking just a massive shockwave killing anything squishy on the planet instantly. Even for the dinosaur there's no D-Day when everyone died. The disappearance of the dinosaurs is a very very very long and gradual period of their declining numbers into extinction. For most of the planet we're talking "just" a climate change. _That_ is what killed the dinosaurs, one way or another. Some species survived that, and in fact even thrived in the new conditions, some species didn't.

Note however there are more hypotheses about that event. The decline in oxygen content in the air in that period, for example, would also be perfectly enough on its own to make a very large beast non-viable. The change in the flora is another candidate. It's entirely possible that the new kinds of plants were either toxic or not nutrient-rich enough for the old lizards.

At any rate, what killed the dinosaurs was _change_. Something changed (take your pick what you think was the killer change there). And some species could deal with it, some species didn't. Dinosaurs (except birds) didn't cope well with the change and their numbers went downhill from there.

Yes, they were a capable species for the old environment, but then the environment changed. And the dinosaurs were suddenly very incapable in the new environment.

So, yes, the dinosaurs are the _perfect_ metaphor for someone or something who can't cope with a change and becomes obsolete.

Change happens. One day you have a nice business hammering scythes and sickles for a village, and the next day someone goes and buys a tractor and a combine harvester and everyone wants _those_. Or you have a nice job calculating tables of numbers by hand and then the CEO goes and buys one of those new "computers". Tough luck. Either you adapt or you're a dinosaur.

It happens with computers and programmers/admins/whatever every day. And some people adapt, some become relics trying to stop progress and return to the good old days. God knows half of the IT departments at big corporations have too many of _those_. Maybe they were once capable and competent. The dinosaurs were too at one point. Now they no longer are. And just like the dinosaurs, sadly it takes a long long time to gradually get rid of those relics. But just like the dinosaurs they _are_ on a slow painful path to extinction.

I'm sorry

[Colin Smith]

The future will eliminate that differentiation. Data will not be 'here' or 'there'. Rather, it will be. Data will simply exist and we will access it as if it were immediately 'here' all the time.

But this is the biggest load of new age bullshit I've heard in years.

Re:power doubles about every two years

[doti]

will that finally be enough to make our computers as smart as we are?

Raw power is not what will make computer as smart as we are.
First, what makes computer "intelligence" is the software, not raw power. And we will need a substantially new software paradigm to get near our intelligence. I can't imagine how software can get consciousness and awareness. There are parts of the human thought that can't be simulated with a series of conditional numeric operations.

"There are parts of the human thought that can't

[phunctor]

be simulated with a series of conditional numeric operations."

You did say it was impossible. You didn't say anything about a new paradigm. Why you'd want to lie about your own publicly visible words totally escapes me.

Still, in case there's a there here: Are you claiming there is a class of problems, such as simulating a thinking human brain, that cannot be executed by a Turing machine? That is an extraordinary claim, and needs extraordinary evidence. Cite?

--
phunctor
"here's a shovel, keep digging"

Re:Losing the difference between here and there

[caluml]

Data will simply exist and we will access it as if it were immediately 'here' all the time.

And precisely where will this data be stored, and how will it get to us? It's not some entity, omnipresent, floating around everywhere, that you can put your hand up, and pull out a load of data.
It has to be stored somewhere. And it has to get from where it's stored to where it's needed.

Re:Losing the difference between here and there

[BadAnalogyGuy]

Where is the data for Yahoo!s servers located? Where is the data for Microsoft's servers located?

Your GMail account's data? Do you know where that is?

No, of course you don't. Because you don't need to. You log in, access the data from the intarweb, fiddle with it, then log off. You aren't doing any of the copying, and the physical location of the data is totally irrelevant for all intents and purposes.

The intartubes are the first step towards removing the requirement of "transferring" data. While some data is pretty well virtualized, a lot of it like files over FTP and large downloads over HTTP are still cumbersome. Beyond that, there is still a differentiation between files on your computer and files on the Web. Why should you need physical access to your computer to access some files? Why do you need to connect via VPN to access those same files from a remote location?

When all things are connected seamlessly you won't ask where the data is stored because it won't matter. What matters is only your ability to access it when you need it.

In the age honored slashdot tradition...

[tttonyyy]

..I'll point out an error in the article.

Replicating Google's 200 petabytes of hard drive capacity would take less than one data center row and consume less than 10 megawatts, about the typical annual usage of a US household.

It's the old rate-of-energy-consumption vs energy-consumed misused once again.

An average household consuming 10 megawatt-hours in a year is pretty dull. An average household consuming 10 megawatts - now that'd be impressive! (Got to power all those gadgets, y'know!)

I think he means that the data center row would consume in an hour the same amount of energy that the average US household consumes in a year.