Handheld Supercomputers in 10-15 Years? 240
An anonymous reader writes "Supercomputers small enough to fit into the palm of your hand are only 10 or 15 years away, according to Professor Michael Zaiser, a researcher at the University of Edinburgh School of Engineering and Electronics.
Zaiser has been researching how tiny nanowires — 1000 times thinner than a human hair — behave when manipulated. Apparently such minuscule wires behave differently under pressure, so it has up until now been impossible to arrange them in tiny microprocessors in a production environment. Zaiser says he's figured out how to make them behave uniformly.
These "tamed" nanowires could go inside microprocessors that could, in turn, go inside PCs, laptops, mobile phones or even supercomputers. And the smaller the wires, the smaller the chip can be.
"If things continue to go the way they have been in the past few decades, then it's 10 years... The human brain is very good at working on microprocessor problems, so I think we are close — 10 years, maybe 15," Zaiser said."
Yes, it will run linux (Score:4, Funny)
Re:Yes, it will run linux (Score:5, Funny)
Re:Yes, it will run linux (Score:5, Funny)
Re: (Score:3, Insightful)
Throwing things on the floor go much faster than 9.8 m/s^2.
With respect to the story at hand. We already have handheld supercomputers.
The Cray 1 was about 100 MFLOPS. Most all cell phones and PDAs CPUs can outperform that.
I work with "supercomputers", and all I see them as are new, expensive, unreliable, and energy inefficient versions of laptops and things.
In the same spirit, some people in the biz call these things time
Re: (Score:2, Insightful)
Throwing things on the floor go much faster than 9.8 m/s^2.
Re:Yes, it will run linux (Score:4, Informative)
Throwing things on the floor go much faster than 9.8 m/s^2.
No it doesn't, at least once the object leaves your hand. Then it's back under the influence of good old gravity, at 9.8 m/s^2, regardless of how fast you may have thrown it.
Re: (Score:2)
But, I'm just an overpaid Microsoft shill sent by them to sow FUD, so what do I know?
Re: (Score:2, Insightful)
Did someone make you say this to stop a terrorist attack? Bruce Willis had to do the same thing in Die Hard 3 by standing in a black neighborhood with a racist sign round his neck...
Re: (Score:2)
Aye, but that's the easy part (Score:5, Interesting)
Now try imagining cooling it. That's the real challenge. That's what makes grown up men cry like little girls.
I mean, look 15 years back in time. That was in 1992. We still had desktop cases without fans (except maybe on the PSU, though even there not on all), CPUs without heatsinks (and in fact, the chip even included in a big slab of resin or such and it made no difference to cooling anyway), and computers could safely run on PSUs whose wattage was a 2 digit number. We also still had RAM fast enough that you didn't need a CPU cache (nor had a transistor budget for it, anyway). And we thought that a program that takes a whole floppy is bloated. Etc.
So I'm going to put on my wizard hat and rub the ol' crystal ball, and tell you how I see computing in the future.
- seein' as case fans started from none, and now we're at two or more 120mm fams and ducts per case, I see the computer of the future as a cube, whose whole face (or maybe side) is one big 14" fan (yes, inch, not cm) blowing air in and another in the back blowing it out. In fact, it will all be one big square wind tunnel, or an oversized hair dryer.
You'll alos be advised to not put anything more flammable than asbestos behind it, and fence it so your cat or toddler can't get behind the computer and get cooked.
- a decent power supply will be around 3-4 kilowatts, but Nvidia will recommend 5 kW for their latest graphics card, more if you run a SLI setup.
- or maybe water cooling will become the standard, and the computer will nicely double as a samovar [wikipedia.org] and espresso machine.
- heatsinks will be made of pure silver. And ATI will still need something that sounds like a jet fighter at takeofff to keep their GPU at only 90C.
- continuing the trend, graphics cards will keep needing increasingly more dedicated power connectors, and increasingly more pins on them. We started at 1 with 4 pins, and now we're at "ATI won't activate this or that function if you don't have 8 pins on the second power connector." I foresee that in 15 years we'll be at 6 power connectors with 16 pins each, just to bring enough current to the graphics card.
- still noone will have invented a better use of all that silicone than adding yet another core, so given that 15 years is no less than 10 cycles of Moore's Law, you'll have anywhere between 2048 and 4096 cores in your PC. More time will be spent passing messages between those and serilizing access to data, in algorithms that were never meant to be massively parallel, than actually computing the useful part. People will still argue that it's the fault of game programmers that they don't split processing 5 NPCs between 2048 CPUs, or for that matter, the fault of compiler makers that they insist on reading the file sequentially instead of each core processing every 2048'th line of the file.
- We'll be up to, oh, maybe DDR9, or maybe some newer standard. It still won't have lower latency in nanoseconds than the old SDR, but people will still buy it based on theoretical burst speed. Even more ridiculously larger caches will be needed just to keep all those cores working at all, instead of spending thousands of cycles waiting for the RAM to finally answer. On the bright side, though, we'll have enough budget of transistors form 2 to 4 gigabytes of cache on the CPU.
- As that trend continues, eventually the disparity between RAM and CPU will get so high that it will be entirely feasible to skip RAM completely, and run the programs off the hard drive and the CPU's L3 cache. (The disparity between CPU speed and RAM latency is _already_ as big as that between the 8088 in the IBM PC/XT and the hard drive it had.)
- People will still take the extra power as an invitation to write bloated and slow code. So even th
Arthur C. Clarke's Steam Cooled Supercomputer (Score:2)
Why supercomputers? (Score:5, Insightful)
Isn't a super-computer a relative term? I mean, I don't know the exact figure but I would that my Dual Core Intel box at home is probably a good deal faster than a super-computer from the 80s. It is probably hundreds of thousands or perhpas millions of times more powerful than the computers used in the Apollo programme. Surely the measure of what is a super-computer and what isn't must be based upon what the fastest machines are in the world at that time.
Perhaps what he means is that what we currently do with supercomputers today will be able to be done with low cost computing. I can certainly see that being true. In fifteen years, it may be possible to adequately simulate nuclear weapons tests, climate models, or protein folding from a run-of-the-mill desktop.
However, the improvements in computing speed will also apply to super-computers. With that extra power you can run more refined models so I can't see how this could obsolete the traditional bulky super-computer.
In short, I can't really understand the super-computer slant of the article. Why not just talk about general-purpose computing instead?
Simon
Re:Why supercomputers? (Score:5, Insightful)
Re: (Score:3, Insightful)
When people don't have news, they make up them. They go and interview anyone who then pulls numbers out of his ass, and thus the "storage technology of the week", "power source of the week", "processing power prediction of the week", etc. is born.
These articles should be considered spam.
Re: (Score:2)
Re:Why supercomputers? (Score:5, Funny)
Re: (Score:2)
Re:Why supercomputers? (Score:5, Funny)
Here are the measurements of my super computer
200,000 Libraries of Congress, or 17 great lakes.
15 Empire state buildings, stacked end to end in a giant circle.
The power consumption of 3 New York Cities.
All the potatoes in Idaho.
Seating for 1.5 747 jumbo jets!
And enough punchcards to circle the moon!
Re:Why supercomputers? (Score:5, Funny)
A little back-of-the-napkin calculation, and we can deduce that if those measurements are equal, then there are 110 bytes per Liter of water.
This makes sense -- if we freeze that Liter, each byte is approximately equivalent to a 1 cm x 3 cm x 3 cm chunk of ice, which I could easily fit into my mouth -- you might even say it's bite-sized.
Re: (Score:2)
Re: (Score:3, Informative)
Also, like most things in computing, "Supercomputer" is a moving target. Today's supercomputers tend to be
Re: (Score:3)
I think the definition of a supercomputer should be changed to something along these line:
"A super computer is any computer which is considered one of the top-N fastest computers in the world today."
Re:Why supercomputers? (Score:5, Interesting)
Yup.
Unless they're talking about something significantly outside the progression we've accepted as Moore's Law. We've come to accept that a super-computer is normally a collection of hundreds of bleeding edge processors. So if they're talking about a handheld ten years from now which is perhaps 1024*(2^(240/18)) times more powerful than a single current bleeding edge CPU, then they could be justified in calling it a super-computer.
They may also be using super-computer to describe a system fast enough that it doesn't need an upgrade to run whatever Carmack pushes out at the time.
c.
Re: (Score:2)
Re: (Score:3, Funny)
No, I think we should insist on a fixed definition of any performance class, which would serve geeks because we could know unambiguously exactly how much computing capacity anybody means when they use a term like "supercomputer". You could even record a conversation and play it back twenty years later, and everybody would know whether we were talking about enough computing power to, say, crack a 56 bit DES key in less than a week.
It would benefit our colleagues in ma
Re: (Score:2)
And we could do it like the guys which gave name to spectra, they shared it in: Low, Medium and High-frequency. Simple. Only, there's not really an upper bound on frequency, now is there ? The result was inevitable.
We then got VHF - VERY high frequency.
Then UHF - ULTRA high frequency.
Then SHF - SUPER high frequency.
Then EHF - EXTREMELY high frequency.
The only thing that prevented us from running into SPHF - Stupendously High Frequency was the fact that by this time, we where running into IR-territory.
Re: (Score:2)
I don't just see that as being "true"
As you already put it, today's PCs ARE super computers relative to the computing power of 10 - 15 years ago. So of course tomorrow's hand-helds will be super computers relative to todays computing power. It's just the way things have gone up until now with no f
Re: (Score:3, Interesting)
$10M or so.
That was the supercomputer of then, and today you can't buy a computer that slow. I don't know what goes in wristwatches these days, but I bet they are fas
Re: (Score:2)
The clock speed of the legendary Cray 1 was 80 MHz. With two instructions going per cycle, you could theoretically get 160 MFLOPS. These are laughable speeds by today's standards, but back then it was considered unbelievable.
Re: (Score:2)
In 1993/4, desktops caught up with this, with the Pentium and PowerPC systems both pushing past this number (Alphas got their earlier, but they were not exactly mass market).
The first handheld chips to push this limit were probably the StrongARM family, in 1995; only two years after t
Re: (Score:2)
Re: (Score:2)
Traditionally, supercomputers were only used to deal with very specific problems which you'd probably write your own software for. They had a lot of very specific hardware designed from the ground upwards for such problems. An algorithm which will get real benefit out of such a system may well perform surprisingly poorly on your dual core laptop.
However, the amount of R&D going into x86 and related architecture has meant that the likes of Cray had trouble keeping up - so many of their lates
Re: (Score:2)
But there has been a general size trend over the last forty years. It's hard to find a computer these days that you can't pick up. Forty years ago a tiny computer was one that could be put on a desk. (And it generally required two people to get it onto the desk.)
Re: (Score:2, Interesting)
Re: (Score:2)
Back then, though, we called them "3d graphics cards".
Re: (Score:2)
Seriously, the definition of "supercomputer" has changed more than once. For example, I vividly remember when personal computers started running afoul of supercomputer export controls [findarticles.com] because they were reaching the astounding speed of 2 gigaflops. Supercomputers didn't reach the teraflop level until the late ni [wikipedia.org]
Re: (Score:2)
From http://en.wikipedia.org/wiki/FLOPS [wikipedia.org]
2007, October: about $0.20 per GFLOPS with the cheapest retail Sony PS3 console, at US$400, t
Re: (Score:2)
Surely you have that the wrong way round? I know for a fact hard drive manufacturers quote sizes in base 10 (so giga is 10^9 not 2^30), and I've been told that bandwidth is the same. RAM is definitely base 2.
You do realise that he means the definition of supercomputers, and not that ones that have already b
Re: (Score:2)
Re: (Score:2)
Did you actually mean it the other way round or are most all credible sources I read on that topic wrong? (Don't mean to be pedantic here, just want to make sure)
The definition of a supercomputer being capable of one trillion FLOPS sounds an awful lot like a series of tubes and an effective copy prevention mec
Re: (Score:2)
Yes. :( I'm having one of those days. I cited the wrong prefix and I got units backwards. I should really be shutting up. Anyway, I've already admitted the other mistake half a dozen times, and I have to wait two minutes between each, which means I can't keep up with the flood of people pointing out the same thing over and over. The first supercomputer was IBM Stret
Re:Why supercomputers? (Score:4, Informative)
Already here (Score:3, Insightful)
Re: (Score:2)
Re: (Score:2)
Nobody cares whose
Re: (Score:2)
Supercomputers from the 1970s are still supercomputers today
No, not by your definition. Neither the Cray-1 (1976, 250MFLOPS) nor the Cray MP-X (1982, 800MFLOPS) count as supercomputers by your definition. In summ
Re: (Score:2)
Since ASCI Red was the first computer with teraflop capability you're saying that there was no such thing as a supercomputer before December 1996?
I don't think so. In the 1970s there was no such thing as a gigaflop computer, much less a teraflop. Perhaps you need to check the timeline [wikipedia.org] for supercomputers on Wikipedia.
Re: (Score:2)
Re: (Score:2)
No, they aren't. A supercomputer can perform a teraflop. That's the definition of supercomputer, and it has been since the word was coined by the government in the 1970s in order to define export restrictions.
Wikipedia [wikipedia.org] doesn't mention that definition. Your claim that the word was coined by the government also disagrees with Wikipedia:
The term "Super Computing" was first used by New York World newspaper in 1929
And Wikipedia actually has a source for that. According to Wikipedia's list, 1 TFLOPS wasn't even reached until 1997, so I can't imagine the United States restricting the export of something that wouldn't exist for another 20 years. I know Wikipedia isn't the final, definitive source for all human knowledge, but until you can provide a source for your information, Wikipedia is mo
The Not Too Far Future (Score:5, Funny)
Re:The Not Too Far Future (Score:5, Interesting)
Re: (Score:2)
Re: (Score:3, Funny)
10-15 years? (Score:2, Redundant)
Re: (Score:3, Informative)
Re: (Score:2)
Re: (Score:2)
Either your first sentence is wrong or your third sentence is wrong. Because everything I can find says that the first computer to reach a teraflop wasn't until 1996 [physorg.com]. So there were no supercomputers until 1996?
Also, the Wikipedia article [wikipedia.org] makes no mention of an official government definition at all, and states that the term was first used by a newspaper in 1929.
Re: (Score:2)
Captain obvious to the rescue! (Score:2, Interesting)
Re: (Score:2)
Hopefully... (Score:2)
Vista slowness -- seriously (Score:4, Insightful)
If you can put a supercomputer in your hand, it's not a supercomputer. A week ago, we had an article here on a guy who'd wired several PS3s together and called it a supercomputer. Folks didn't agree with the supercomputer designation, even though he was getting flops that would clearly have been supercomputer speed just five or six years ago. It's not speed that defines a supercomputer, it's speed relative to what's commonly available.
If we crunch down machines to incredibly small size, then research institutions will buy one 50 times that size. Every time. What will happen is that that tech (if it's not expensive) will drive PC speeds up, perhaps phenomenally, software development tools will make use of the extra speed to make programming easier at the expense of run-time, and we won't see significant speed increases in the user experience. The user will be able to do more, of course, but he'll be complaining "When I speak into the microphone to tell it to write a three page synopsis of this book in it's library, it stalls and lags, and sometimes I tell it twice, before I get a response, and then it gives me two outputs. This thing is SLOW."
Re: (Score:2)
Yeah, you're just dead wrong, here. That guy is a professor of computer science at North Carolina Sate; his name is Frank Mueller. And, surprise surprise, he knows comp sci better than you do. "Supercomputer" is a legal term coined in the 1970s by the US government to define export restrictions on computing hardware. It has a c
Re: (Score:2)
We already have handheld supercomputers (Score:4, Interesting)
No, really. An iPhone is much more powerful than the Cray-1, and probably significantly more powerful than a Cray X-MP. The iPhone certainly has much more RAM and storage than they typical early Crays; I can’t be bothered right now to find out what kind of MFLOP performance an iPhone has.
Cheers,
b&
Re: (Score:3, Informative)
I'm not sure why you believe this. I'll assume you mean the Cray 1A, since the Cray 1 is just a specification; it's a bit like talking about the 386, since the 386 ran at about a dozen different clock speeds. The Cray 1A was the first actual implementation of the Cray 1 spec, and was initially installed at Los Alamos. SCD's Cray 1 was installed about six months later, and ran at 160 mega [ucar.edu]
Re: (Score:2)
Do your homework.
Re: (Score:2)
They generally have wider memory buses, lower memory and network latency, etc... etc... designed into them.
Re: (Score:2)
Except that most supercomputers on the Top 500 list aren't defined as such because of their raw memory, or MFLOPS, etc... Supercomputers are different from the average PC/iPhone/whatever consumer device not quantatively - but qualatively. Not of degree, but of kind.
Which makes the appellation "supercomputer" even LESS appropriate for this device, wouldn't you say?
Nonsense (Score:5, Funny)
Am I missing something? (Score:5, Insightful)
The analogy I've seen comparing big iron midrange and mainframes vs. PC's is "Yeah, the PC is zippy, but it's like a ninja bike. The big iron is like a dump truck. The midrange isn't going to get up to speed as quickly but it's going to be doing a hell of a lot more for the effort."
Re: (Score:2)
No. Supercomputer is a specific FLOPS threshhold established by the government in the 1970s as a basis for export restrictions on hardware. A supercomputer from 1972 is a supercomputer today. It has nothing to do with generational standards; otherwise, stapling 8 PS3s together wouldn't prove anything, and it would be impossible to ever get a supercomputer in one's hand, given that the standards of the day would be
Re: (Score:2)
http://mathstat.asu.edu/support/doc/unix/coping-with-unix/node188.html [asu.edu]
supercomputer: The class of fastest and most powerful computers available.
As for the US government's export regulations - the definition of a High Performance Computer (HPC) was raised from 28,000 millions of theoretical operations per second (MTOPS) to 190,000 MTOPS on December 10, 2003.
http://www.bis.doc.gov/hpcs/ArchivedNewsItems.html [doc.gov]
No handheld supercomputers (Score:3, Interesting)
Power (Score:2)
Extrapolating power consumption over the last ten years would seem to indicate that this "super computer in your hand" would probably be glowing red hot. Before we increase computing power much more we need to get a handle on efficiency.
Re: (Score:2)
Oh, bullshit. There are several laptops on the open market right now that are over the 50% line. The only palmtops that glow red hot will be doing so because they use Sony batteries. Saying things like "extrapolating" without actually doing the math really just makes you look like an asshole. You aren't extrapolating. You're guessing.
Re: (Score:2)
10 -15 years away 50 years from now.... (Score:2)
Been there, done that. (Score:2)
Re: (Score:2)
*POOF* (Score:4, Funny)
We have hand held supercomputers now (Score:2)
=mcgrew
Great in Winter (Score:2)
TWW
Make it stop (Score:2)
To be fair... (Score:2)
Pedantically speaking (Score:2)
Re: (Score:2)
handhelds vs supercomputer benchmarks (Score:3, Interesting)
The Nokia has 64 MB RAM. The '205 had 16 MB RAM. The Nokia kicks scaler code at about 40 to 100 MIPS. The '205 kicked scaler code at 35 to 70 MIPS. The Nokia has a DSP, which seems to be able to kick about 200 MFLOPS (i could be wrong). The '205 had twin vector pipes with a peak performance of 200 MFLOPS each, but it was rare to get more than 80% of that. My point is that they're comparable. The Nokia came with 192 MB file store, but now has 2.1 GB, and can mount my desktop filesystems over WiFi with better than 1 MB/sec throughput. The '205 had about 1 GB disk, and could mount mag tapes. Both sport native compilers for C, Fortran, etc. The Nokia was about $150. The '205 was about $15,000,000. That's a factor of 100,000 improvement in price/performance. The Nokia runs on batteries and fits in my shirt pocket, with the form factor of my old Palm Pilot. The '205 had a motor-generator power conditioner (the flywheel acts like a battery in power failure) and fit in large machine room with temperature and humidity carefully controlled.
Would i call the Nokia a supercomputer? No. Supercomputers cost more than a million dollars when they are new. Would i build a beowulf cluster of Nokia's? Maybe. With WiFi, one might put together an ad-hoc grid pretty easily. I only have one. But my 4 year old desktop is more than 30 times faster, so it's going to be hard to justify from a pure performance standpoint. Yes, my desktop has better price/performance than the Nokia.
I've not yet run a SETI@Home unit on the Nokia. It'd be much better than the one i ran on the 486/33...
The perfect stocking stuffer (Score:2)
A we got them B we will never have them (Score:2)
What is correct? Both.
By alrights todays average computers ARE supercomputers, you just have to measure them against the supercomputers of the past.
By that same token you will NEVER have a handheld superocmputer because by simply combining a couple of them together you would have an ever more powerfull computer.
So the article basiclly states, in the future you will have more processing power then you have today. Mmm, yeah, that might happen.
...but... (Score:2)
Hell, my palm pilot way outpowers my CoCo II. Are any of you OLD ENOUGH to remember the CoCo??
happened several years ago (Score:2)
w far we've come (Score:2)
Phone
CPU: PXA272 312 MHz
memory: 64 mb
Drive: 64 mb built in flash 1gb mini-SD card
First comp
CPU: 486 dx 33mhz
Memory: 8 mb ram
drive: 350 mb HD
It's not a huge leap to assume todays desktop will be tomorrows mobile device 20 years down the line.
Weird definition of "supercomputer" (Score:2)
He's got a pretty bizarre definition of "supercomputer." I've always understood supercomputers to be the fastest, craziest computers currently available. Obviously, this changes over time. I propose that a hand-held computer BY DEFINITION cannot be a "supercomputer." It may be a very, very fast computer. But take thousands of such hand-held "supercomputers" and slap them together, that's a REAL supercomputer. Just like it's always been.
In 1980, many of our desktop machines would have been considered "supe
Re: (Score:2)
Re: (Score:2)
What are they gonna call the ubuntu that comes after the Zesty Zephyr?
Re: (Score:2)
What are they gonna call the ubuntu that comes after the Zesty Zephyr?
Didn't fully convince me either.
Re: (Score:2)
Not to mention heat dissipation.
-:sigma.SB
Re: (Score:2)
Re: (Score:2)
I'll mail you a dollar if you do that. The PS3 is eleven pounds; it'd be hard enough to hold up fifteen pounds as a stack of eight unbalanced pieces of that size, one on top of the next, and around half of people couldn't hold 88lbs with a single hand. Holding up an unbalanced stack of 88 lbs would be a serious feat of balance.
And remember, if you try it with an XBox, you're not even going to be able to lift two controllers...