Slashdot Mirror
China Makes World's Fastest Supercomputer
shmG writes "China has replaced the United States as the maker of the world's fastest supercomputer. A Chinese research center has made the world's faster super computer — named as Tianhe-1A, which was released at a national conference on high-performance computers (HPC) in China. Made at a cost of over $88 million, Tianhe-1A is theoretically able to do more than 1 quadrillion calculations per second (one petaflop) at peak speed. Tianhe-1A 's peak performance reaches 1.206 petaflops, and it runs at 563.1 teraflops (1,000 teraflops is equal to one petaflop) on the Linpack benchmark."
But does it run (Red Flag) Linux?
Maybe this is the bitch-slapping the US needs to pull it's head out of it's ass, and start doing the things it needs to do to be seriously competitive again.
Meh. Who am I kidding?
American Third Position
Finally, a real choice!
And the fastest social and economic downturn is in America...coincidence?
He who knows best knows how little he knows. - Thomas Jefferson
Stolen? I don't know. Purchased? From the article:
So unless Nvidia and Intel have reported 20,000 or so stolen processors lately, I wouldn't worry too much.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
This is what the rukept telling themselves during the Cold War.
Serious research still needs much faster supercomputers than we have now. All kinds of science - from artificial intelligence to weather modeling to astrophysics to genetic research to nuclear simulations. Access to a powerful supercomputer is a major boon for academia in the country.
Or they're using them for complex modelling - like everyone else who has one uses them for.
If your definition of supercomputer is an old yellow box with the turbo button then yes. Otherwise you are just plain retarded.
As someone whose works depends on HPC I disagree with you. A lot of people in life sciences, materials science, nuclear physics, geophysics and other knowledge areas needs clusters and super computers.
English is not my first language. Corrections and suggestions are welcome.
Finally they will be able to computerize the national census procedures.
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
So far the numbers have not been independently verified, correct?
I mean, who ever expected that Skynet would speak Chinese?
Never underestimate the potential of Human stupidity. -Heinlein
The battle to build the fastest supercomputer has become a sort of national pride as these high performance computers are used in a variety of areas like defense, energy, finance, science. They are also used for drug discovery, hurricane and tsunami modeling, cancer research, car design, even studying the formation of galaxies. For example, oil companies use supercomputers to find reservoirs, while Wall Street traders use it for speedy automated trades.
Everyone knows that you can do all of the above on a simple netbook
After observing Red Flag, Loongson, and the basic nature of Chinese hardware, I predict we're going to shortly see an "oh wait, they were lying, it's 200 teraflops of American hardware" come down the pipe.
I wouldn't mind being wrong, though.
Oak Ridge (Jaguar):
Cores Rmax(GFlops) Rpeak(GFlops) Nmax Nhalf
224162 1759000 2331000 5474272 0
Seems faster by a good margin.
we went to the moon during a period of nationalist chest thumping, and when the nationalist chest thumping subsided, we haven't been back. countries that are interested in nationalist chest thumping: china, india, etc, are still pumping up their space programs
what i am saying is, for all the evils of nationalism, scientific advancement in the realm of large projects seems to be a positive byproduct
for example, if we were still in a cold war with the ussr in the 1990s, i will bet you anything that this would have been completed and would be producing amazing science at this point in time:
http://en.wikipedia.org/wiki/Superconducting_Super_Collider
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
The slashdot summary has the wrong numbers. The actual article which slashdot quotes is contradictory. Its starts by saying:
"Tianhe-1A has set a new performance record of 2.507 petaflops, as measured by the Linpack benchmark, making it the fastest system in China and in the world today."
and then one paragraph later it gives the same numbers as the slashdot summary.
Other articles (from other sites) are claiming theoretical peak performance of 4 Petaflops (from an Nvidia source) and sustained petaflops of 2.5.
As a researcher using (US&European) supercomputers for high-performance numbercrunching, I strongly disagree. Some of the calculations take so much time that if I had to run it on one single (powerfull) CPU, I could probably do one big calculation in my entire PhD... Nevermind that by the time it was finished, the hardware would be long obsolete, and its not possible to fit the data in a single computer's memory anyway. So for some tasks, large/powerfull & shared computer resources really makes sense.
But of course, computers, electronic calculators, abacus'es, and pen&paper isn't strictly neccessary. Get off my stone-garden so I can carve my calculations on it!
So apparently all it takes to make a "supercomputer" now is lots and lots of money. Just go to Newegg and order that dual socket xeon dream machine you have in your cart. Then order 14,000 more of them.
From TFA: ... Tianhe-1A is theorectically able to do more than 1 quadrillion calculations per second(one petaflop) at peak speed. Tianhe-1A's peak performance reaches 1.206 petaflops...
"... Tianhe-1A has set a new performance record of 2.507 pataflops, as mesured by the Linpack benchmark
So does it do 1, 1.206 or 2.507 petaflops ?
There is absolutely no consistency in numbers in this story. Some measurements show this computer to be about 45% slower than the Cray XT5 and some show it to be faster. Given China's history of arbitrarily throwing out numbers to try to prop themselves up in the international community I cannot accept this as fact without some sort of independent verification. If China has in fact created the worlds fastest supercomputer, I congratulate them on a job well done. But I am still skeptical about this story. Sounds like my government (US) is just looking for an excuse to spend billions more on a new supercomputer.
Watch it... he has a botnet of Atari 2600's that will DDOS your ass.
Or something.
Remember to maintain your supply of
Is this faster than Blue Waters at NCSA is going to be in 2011?
The article starts by claiming 2.507 petaflops, but gives no mention if that is Rmax or Rpeak. We have to assume that it is Rmax, since 2.5 petaflops is no big deal in terms of Rpeak.
Unfortunately, then the article lists both Rpeak and Rmax. But the numbers quoted seem to be for Tianhe-I (#7 on the top 500 list), not Tianhe-IA (not currently listed). Wikipedia table of the top 10
Oh, and it gets better. The article claims that Tianhe-IA has 7,168 GPUs and 14,336 CPUs. Very strange, since the Tianhe-I has 71,680 CPU/GPU pairs.
My guess is that China doubled up their Tianhe-I computer and swapped out for newer GPUs, then named the new thing Tianhe-IA (this is pretty normal when competing for top500 spots). I'm going to go with 143,360 Xeon/M2050 pairs. Either that, or the Chinese found a way to overclock 10% of their chips into the 20+ GHz range and threw out the rest.
See that "Preview" button?
INTEL, IBM, and other high tech firms have been sending their R&D, engineering and other high up on the job food chain jobs over there and to India. They have been building up expertise in other countries. Of course this happened.
We the US will become a technological backwater. Of course the pundits will say shit like "American kids just aren't studying science and engineering" or "It's our education system."
The answer is: why should a bright kid go into science or engineering when he won't be able to get a job? Whereas, if he goes into medical, he's pretty much guaranteed a very nice living.
It's not the education system; it's the market. The market here in the US is saying that engineering and science careers just aren't worth as much as others and it's saying that there are plenty of qualified and cheaper engineers overseas - all thanks to US companies moving there.
As we are seeing NOW, the Chinese and Indians no longer need American companies - they don't need IBM or whoever to come in a spend the millions setting up shop. They can do that themselves now thank you very much. End result: US based companies will be sidelined.
So kids, apply to foreign firms because US based companies have made themselves irrelevant.
And business owners, bypass the middlemen (IBM and whatnot) and buy direct from their suppliers in India and China - you'll save the costs of over paid American management and sales people.
RIP America
July 4, 1776 - September 11, 2001
To be fair, it is not like the US government is not also involved in corporate espionage:
http://www.europarl.europa.eu/sides/getDoc.do;jsessionid=392426F8AEF3E495934CBA3FB0007F30.node1?pubRef=-//EP//TEXT+REPORT+A5-2001-0264+0+NOT+XML+V0//EN&language=EN
Palm trees and 8
Does research still need supercomputers, though? If you're writing a program parallelized enough to split across tens of thousands of processing units, why not go with a full cluster, like the vaunted Hadoop or EC2?
I think the time for a single powerful machine is long past. Maintaining the level of interconnection between the nodes is expensive, and we can do better. In the words of Dr. Ken Batcher, "A supercomputer is a device for turning compute-bound problems into I/O-bound problems." With distributed storage (like HDFS) coupled to the distributed processing (like Hadoop), we can turn those same problems into merely cost-bound problems.
You do not have a moral or legal right to do absolutely anything you want.
Supercomputing is largely a solved problem...do a regression analysis on the variables on the Top 500 sometime. Primarily it's a function of the number of cores. I've told people in my own workplace - you want a machine on the Top 500 then write me a cheque - making a supercomputer isn't the feat of skill or engineering that it was in the days of Cray. This doesn't even touch on what the hell you use these things for the problem space for parallel processing is clearly smaller than that of serial processing. Add to that the assertion that the number of useful applications drops off steeply as you add more cores (at some point you are left with only the "embarrassingly parallel" ones) and creating the largest supercomputer in the world is akin to saying you are creating the least useful computer in the world. Not to mention probably one of the least power efficient, highest maintenance costs, etc..
What metric should one use to compare supercomputers? Using the amount of floating point operations performed per second (FLOPS) comes across as a little silly, because if you just pile together a sufficient amount of standard PC's, you should be able to top it (maybe Folding@home could be considered a bigger supercomputer). Therefore, I assume that the interconnectivity (and the speed related to it) of the CPU's should have something to do with it...
If Pandora's box is destined to be opened, *I* want to be the one to open it.
There will always be problems that require large amounts of computing power. In some cases, increases in computing power make previously unworkable problems feasible to throw a computer at.
Just because today's PCs are more powerful than older supercomputers doesn't mean there is going to be demand for capabilities at the upper end of the computing power spectrum.
In this case - A large number of the top computers in the world are used for nuclear weapons simulations. (You can't test them any more due to test ban treaties, so you have to simulate them.) So China ramping up their computing power is a bit scary. Note that it was apparently created by a "University for Defense Technology" of some sort.
(Note: I can't read TFA, as it appears to be trying to give me a popup ad, but the big square covering the article is blank and has no close button. However I've seen one or two smaller articles regarding this new system.)
retrorocket.o not found, launch anyway?
Oh, like this doesn't go both ways. At least they didn't start bribing/kidnapping our scientists to make "Dooms Day" devices in a hidden lair in the desert, like we did with the Germans in the Manhattan project.
Also that Washington Times article is way out of date. The proceedings of the Wen Hoo Lee trial (accused of stealing the W88 warhead for China) embarrassed public officials to the point where Bill Clinton had to issue an apology and the US government had to pay an undisclosed settlement to Lee.
The fact was US military research bases had slack handling protocols for classified documents. This was glaringly obvious as the cause of all the leaked secrets. However, to save their careers, the managers made Lee a scapegoat. Turns out it was common practice to take classified documents to work on at home. The only reason Lee was singled out for it was, because he was Asian.
Nearly all of these supercomputers are just that - VERY large clusters.
Although in many cases they have specialized communications backplanes for communications between nodes with capabilities (such as low latency) that can't be achieved with geographically distributed clusters. (Note the mention of parts from Intel and Nvidia, combined with undefined "domestic" communications silicon.)
Also note that geographic distribution leads to all sorts of information assurance nightmares when you're simulating nukes...
retrorocket.o not found, launch anyway?
The only reason it's so fast is because a half hour after you feed it data, it's hungry again...
There exists no way of exchanging information without making judgments. --Bene Gesserit Axiom
- what "nationalist chest thumping" went on to make the Large Hadron Collider happen?
just curious to see how this fits into your theory. I've no idea, maybe you have the answer. But the LHC seems to have got built, and not as a war time artefact (in my ignorant opinion). Seems more like a collaboration between nations.
Education on this point welcomed.....
"Does research still need supercomputers, though?"
Yes, lots of good, interesting research still going on.
If you think clusters or Hadoop or map reduce or EC2 or cloud computing is going to solve the problems that supercomuter tackle you will be waiting around for a very long time.
"A supercomputer is a device for turning compute-bound problems into I/O-bound problems."Some problems are compute bound, some are memory bound, some are I/O bound. Supercomputers attempt to addresses this for the largest of these types of problems. Distributed storage or distributed process like Hadoop does not solve this except for the most trivial of supercomputer-class problem. There is still a need for supercomputers.
his sources. As reported on CNN from Mashable:
Unveiled Wednesday at the Annual Meeting of National High Performance Computing (HPC China 2010) in Beijing, Tianhe-1A is the world's fastest supercomputer with a performance record of 2.507 petaflops, as measured by the LINPACK benchmark.
Tianhe-1A was designed by the National University of Defense Technology (NUDT) in China, and it is already fully operational.
To achieve the new performance record, Tianhe-1A uses 7,168 Nvidia Tesla M2050 GPUs and 14,336 Intel Xeon CPUs.
It cost $88 million; its 103 cabinets weigh 155 tons, and the entire system consumes 4.04 megawatts of electricity.
Tianhe-1A ousted the previous record holder, Cray XT5 Jaguar, which is used by the U.S. National Center for Computational Sciences at Oak Ridge National Laboratories.
It is powered by 224,162 Opteron CPUs and achieves a performance record of 1.75 petaflops.
According to Nvidia, Tianhe-1A will be operated as an open access system to use for large scale scientific computations.
Just sayin...
Hey KID! Yeah you, get the fuck off my lawn!
China: "We made the fastest super-computer!!!"
Intel and NVidia: "Uh - no you didn't, We are own all your processors!!!"
IBM and the NSF are already building a supercomputer which will have an eventual supposed peak performance of 10 petaflops: Blue Waters http://en.wikipedia.org/wiki/Blue_Waters At the university of Illinois. I actually work at the university and have been by the building (it is complete), though we don't have the hardware yet from IBM. It is all starting to go in, though, and is supposed to be working before October of next year. So don't worry, it's not like the US is getting stomped in computing power.
Also, to those posters questioning the need for supercomputers......as someone who works on HPC code and applications, I cannot disagree with you more. We already have simulations that are not practical to run on modern supercomputers. In the grand scheme of things, these are not even terribly "complex" issues when compared to, say, modeling an entire human organ or even body. Imagine what we could do with something like that....you could test new "drugs" in a simulated environment and get a good idea of the effects (this is all just one idea that comes to mind)...
well if its a war, lets start fighting back
implicit in your statement is that citizens are somehow helpless against corporations. hardly
don't lay down and take it, stand up and fight back, if at least only to preserve your self-respect
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
Germans on the Manhattan project wat? I think you mean building rockets and leading NASA
in about a year that it's full of lead, heavy metals, and other chemicals that are bad for babies and pregnant mothers.
Due to the Von Neuman bottleneck, most of the transistors in a computer are in the RAM, which is idle except for the row/column being accessed at a given time. The bitgrid gets around this by building a grid of look up tables which operate on 4 bits in and 4 bits. It should be quite easy to build a chip which has a million of these tables in a 1000x1000 grid. This would allow data to flow off the edges at a result per clock cycle.... which in modern CMOS is at least 1 Ghz.
Finding applications to run on a theoretical chip isn't easy... but synthetic focus imagery comes to mind. Imagine a survey plane with an array of cameras, generating 3d imagery with a resolution of 1 centimeter in all 3 dimensions, at a speed of 1000 kph at a height of 15,000 meters, giving a swath 5 km wide.... that's 14 Gigapixels/second of final product. I believe it's feasible to do this with the bit grid chip.
Anyone want to sponsor the research? If not, I'll stick to my day job.
Call me when China invents something like Tang.
The real Sig captains the Northwestern. This one captains
I'll bet it can't!
if this is supposed to be a new economy, how come they still want my old fashioned money?
China: "We made the fastest super-computer!!!"
Intel and NVidia: "Uh - no you didn't, We are own all your processors!!!"
The magic behind supercomputing isn't CPUs.
The real trick has always been the interconnects & the software that gets those thousands of C/GPUs talking to each other.
And don't ever forget that China is developing its own CPU.
It still sucks right now, but like with rare earth minerals, China is in this for the long haul.
[Fuck Beta]
o0t!
Hyperinflation sure sounds nice to most people who have large student loans.
With the creation of the worlds fastest computer China no longer will be thought of as the land of cheap labor, but also as a land inhabited by the worlds fastest thinkers. Research and development was once monopolized by the US and it was the leader in the race for new products, but a yellow horse has now taken the lead. Unless the US regains its once creative workforce, it will finish out of the money.
Yes, this is spot on for massively parallel systems. The interesting thing is that China does actually make their own interconnects, but they aren't so great. The Tianhe-1 actually runs at 47% of the theoretical capacity. In contrast, the previous number 1 (Jaguar) runs at about 76%. In fact, China's previous big HPC was Nebulae, which had a higher theoretical peak than Jaguar, but didn't actually perform faster because of interconnects problem.
Gentlemen! You can't fight in here, this is the war room!
"the costs of over paid American management and sales people."
and there's the key. the U.S. is the victim of its own isolated success. All of a sudden its no longer isolated (a la The World Is Flat), and the jobs move to cheaper venues.
I did just read an article about certain aspects of Indian cost of living increasing in accordance with the shift in jobs. We just aren't worth what we say we are, and the market correction is taking a long time.
The time for "supercomputers" is long past. This is just muscle-play that does not mean anything. The fetish of having a "supercomputer" seems to be a left-over from the times when computers were very slow and almost nobody had one. Or maybe politicians (and journalists) are still living in those times...
False. 1) Protien Folding Simulations (and any number of chemical / biological simulations). 2) Prestige attracts money and brainpower.
Stunt, yes. You could make several smaller "super computers" to have equivalent output. Worthless no.
postmodernsideshow.com
After an hour of processing I contracted lead poisoning and was inexplicably hungry...
Serious research still needs much faster supercomputers than we have now. All kinds of science - from artificial intelligence to weather modeling to astrophysics to genetic research to nuclear simulations. Access to a powerful supercomputer is a major boon for academia in the country.
Whether serious research needs faster computers, is up for debate. Personally I think it mostly needs researchers that know how to program. But even if faster computers are needed, then the needed ones are not of the supercomputer type.
I hope you realize that "FLOPS" are entirely useless for AI. On the other front, these "supercomputers" give you very little "FLOP" for the buck.
It is not a boon. It is a waste of money, done in order for some people to build themselves a monument.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
And to add to that, if you are I/O bound, you are screwed anyways, and even more so with this "supercomputer", as it has an overkill of local processing power in the nodes that will easily saturate any backplane. Graphics-card based computations make only sense if you are _not_ I/O bound. But in this case a cloud is far better than one monolithic system.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Depends on the task at hand. If something is infinitely parallizable such as a ray-trace where a screen can have subsets of pixels be handed off to other cores, CPUs, or separate discrete computers, just chucking more blades in a rack will do the job.
But not all tasks can be broken up into bits that don't depend on each other. There are tasks such as various types of modeling which require step "A" to be done before step "B" can be handed out, and even with the best of technology, if a box had 1000 cores, these tasks would only run as fast as one lone core.
Super computers (just like mainframes) still have their place. They are not as awesome as in the past because of the compute power available to the average person, but they are still needed.
You have no idea what you're talking about. There will always be a need for lots of computational power connected by networks with lots of bandwidth. A supercomputer isn't "super" because of the flops. It's "super" because of the network and the software.
The issue is I/O. CPU power is one thing. However, getting the data up the storage hierarchy to the CPU and back down again is where people pay the big bucks for real machines and not just fire up stacks of x86 boxes if they have some serious tasks.
This can be explained in a simple way: Build the latest Linux or BSD kernel. The time it takes to build one either has stayed the same, or actually has gotten longer than in times past (when one ran a kernel build of that time on that time's computers). Why is that, even though CPU power should be going up exponentially? I/O is not improving as fast as CPU numbers. It is improving, but it isn't doubling as often. So, even though the CPU is snappy, it still sits there waiting for the fetches from disk, to RAM, finally to the registers. Cache helps ease this pain, but I/O is still the bottleneck for a lot of tasks. If this was not the case, it would take less than a second to rebuild a kernel from a make clean.
CPU is cheap; I/O is expensive, relatively.
No. Please just stop speculating. Supercomputing is not driven by the TOP500. It's driven by agencies that do real scientific research. We're not pumping out flops and GB/s to make the TOP500. We're doing it because our scientists need it.
This rings quite true, although it is slowly changing. Here in the US, you can get out of college with a fresh CS or scientific degree from a good university... and end up sitting on your duff for years waiting tables until you find something relevant. A simialr student who finishes up a generic major in college, then passes the bar exam in their state, will never see an unemployment line in their lifetime.
Until this is changed, Americans will see the cool electronic stuff only appearing in China and India first, just like the cool smartphones end up only in Japan or Korea and never make their way across the pond. Same with Internet stuff. Bandwidth in most parts of the US is actually shrinking, while in virtually every other nation, companies are busy laying fiber or putting up wireless towers.
Tianhe-1A is theoretically able to do more than 1 quadrillion calculations per second (one petaflop) at peak speed. Tianhe-1A 's peak performance reaches 1.206 petaflops,
But according to the news Jaguar is able to mosey along at 1.75 petaflops, and Tianhe is rolling along at 2.5. The petaflops barrier was broken a few years ago, so Tianhe is basically a product of Moore's law.
Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
The 3 gorges will have nothing on us. That'll show 'em!
And if destroying the natural beauty isn't enough, I'm sure we can find some priceless pueblo or ancient burial mound, and turn it into a cooling pond for a giant nuclear plant. Take that, China!
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Citation please. The only reason we can't meet these obligations right now is that the Republicans and their rich cronies don't want to pay their fair share into the system.
Replying to a massive troll, but what the heck.
Nowhere did I say raising taxes was the sole solution. We do need to raise revenue and we need to raise it from those who've been living off the hard work of the poor and (rapidly shrinking) middle class. We also need to rip apart a lot of existing systems and rebuild them from the ground up. Not necessarily because they're inefficient but because they're aimed at the wrong goals. Pumping money into a system that's oriented the wrong way isn't going to help anything.
This is not a Democratic or Republican thing. It's common sense. I don't share your cynicism about the supposed lack of intelligence of the electorate. It's simply not very easy to know about the political intricacies when you're working three jobs and have a family to take care of.
That said, it is absolutely false that both parties are equally to blame. The Republicans are much more in bed with big business and generally advocate policies that humanize corporations (see Citizens United) and dehumanize human beings (see opposition to health care access). There are bad players on both sides but one side has many more than the other and the official planks of one side actively encourage that behavior.
Social Security would be solvent if we raised the limit on taxable income. Why is there even a limit?
Our whole health care system would work much better if we went to a single-payer system or to any of the other systems where health is treated as a public good. This has been demonstrated across the world. Simply look at the health outcome numbers and the funding input numbers.
It's really, really not rocket science to fix most of our major problems. The problem, as always, is political will.
To put it in perspective:
Chinas new super computer:
$ 88,000,000
US defense budget 2009:
$ 515,400,000,000
Yeah, I can't see how you could possible be able to afford anything like this!
(And who knows, eventually your military already has it, or much better gear, just not bragging about it in the open.)
Yeah, this supercomputer is totally 'meh', at reading the title I was hoping they used https://secure.wikimedia.org/wikipedia/en/wiki/Loongson#Loongson_3 not nvidious crap.