Japan Builds World's Fastest Computer 549
claylikethemud writes "The New York Times reports that Japan has built the world's most powerful supercomputer from "640 specialized nodes that are in turn composed of 5,104" NEC processors. The machine boasts the computing power equivalent to the 20 fastest American supercomputers combined, and with a top speed of 35.6 teraflops, outpaces the next fastest machine, the ASCI White Pacific, by more than factor of five. Applications include climate modeling, global warming prediction, and other non-weapons research."
Did anyone see this coming? (Score:5, Insightful)
Re:Did anyone see this coming? (Score:5, Insightful)
For some reason Japan isn't all that keen on nuclear weapons.
--
Evan
Re:Did anyone see this coming? (Score:4, Informative)
Or is it? A Japanese no longer unthinkable [yahoo.com]
Re:Did anyone see this coming? (Score:2, Funny)
No, they arn't they seem to prefer the less traditional giant robot to the nuclear weapon standard.
I wonder why...
Re:Did anyone see this coming? (Score:2)
Re:You do not know what you are talking about (Score:3, Insightful)
Nuclear weapons are the most sensitive issue in Japan, Japanese people are strongly against it. Since the nuclear accident in Ibaraki Prefecture in 1999, the most serious nuclear leakage accident, Japanese citizens have lost confidence about nuclear industry, they asked governments to reduce or stop nuclear power plant construction.
So how, exactly, do I "not know what I'm talking about"? --
Evan
Re:Japan and weapons. (Score:3, Interesting)
--
Evan
Re:Did anyone see this coming? (Score:2)
Re:Did anyone see this coming? (Score:2, Insightful)
Re:Did anyone see this coming? (Score:2)
I'm pretty sure this is where I heard about it months ago.
Read about it here (Score:5, Informative)
Info. you don't need to subscribe for here:
http://www.bayarea.com/mld/mercurynews/business
here's a link that works (Score:2)
The above link has a space in it and it's not a link.
PS Don't bother modding this up, I'm already pinned at 50.
Powerful supercomputer (Score:5, Funny)
Re:Powerful supercomputer (Score:5, Funny)
US:bombs vs. Japan: environment (Score:3, Insightful)
The accomplishment is also a dramatic statement of contrasting scientific and technology priorities in the United States and Japan. The Japanese machine was built to analyze climate change, including global warming, as well as weather and earthquake patterns. The United States has predominantly focused its efforts on building powerful computers for simulating weapons.
Re:US:bombs vs. Japan: environment (Score:3, Interesting)
Not surprising. Not in the least. Of course, the United States government is going to be spending its cash on simulating nuclear weapons. They have to. They have nuclear weapons. The Japanese, as far as I know, don't. Japan was stripped of a military after World War II probably because the U.S. feared that the same thing would happen to Japan that happened to Germany after WWI, that is, that Japan would get strong again and attack. That is probably why they don't have a military (I think they now have a token military but not a real one) and have no need to simulate nuclear weapons.
As for being the fastest. IBM's Blue Gene [ibm.com] will outstrip this Japanese model in two or three years. That's the downside. It will be two or three years from now. Oh well, it will run at One-Petaflop.
Re:US:bombs vs. Japan: environment (Score:5, Insightful)
Actually, Japan has one of the largest military budgets in the world. They call their military the `Self Defense Forces', but it's the real thing, with big ships, tanks, fighter jets, and all that good stuff. No nukes though.
Re:US:bombs vs. Japan: environment (Score:3, Informative)
The US Budget is 265 Billion.
My problem with what you said, is that you made it seem that Japans budget is close to the US's budget.
All in all, it is a surprisingly large amount for a country that doesn't go into military actions. Who are they defending themselves from?
Do you really have to ask? (Score:2, Insightful)
Red China and North Korea, for starters (who both have nukes, BTW). They don't exactly have the friendliest of neighbors over there. They would be stupid not to have a good defensive force.
Re:US:bombs vs. Japan: environment (Score:5, Funny)
Re:US:bombs vs. Japan: environment (Score:4, Insightful)
The US ideally would go to war where only weapons ould be destroyed and noone would get killed.
To acieve this goal our weapons have to be extremely high tech.
Oddly enough it's value of all human life, both ours and the people in the region we're fighting that makes our budget so huge.
Cheers,
Jonathan
Re:US:bombs vs. Japan: environment (Score:2, Insightful)
Yeah, sure. Keep thinking that.
Re:US:bombs vs. Japan: environment (Score:2)
Re:US:bombs vs. Japan: environment (Score:4, Insightful)
You never know when some pesky Canadians on a training exercise in Afganistan will suddenly turn their weapons on an American F16.
Re:US:bombs vs. Japan: environment (Score:2)
I know this may sound like a joke but the military did turn around and perform a study of how to defend the U.S. against Canada right after WWII, I think, mostly because there was nothing to do (all the bad guys had been defeated).
Really, I think, the exercise of how to defend against an attack by Canada was more for practice than for anything else.
Re:US:bombs vs. Japan: environment (Score:2)
My problem with what you said, is that you made it seem that Japans budget is close to the US's budget.
Japan's budget is big enough that the US wouldn't attack them. Israel spends much less on it's military than France, but they can fly twice as many sorties than the US in the middle east. If they didn't have that military we would probably have sent in UN troops a long time ago to make the palestinian concentration camps more livable. (I mean concentrarion camp in the traditional way, not as the WWII euphamism for the death camps.)
I'm one for a stronger Japanese army, it's already strong in a conventional sense, but nuclear weapons would be a plus if only to keep an future war conventional. They have a vibrant democracy, and I don't see that changing even if they are experiencing their "great recession."
Re:US:bombs vs. Japan: environment (Score:3, Interesting)
Now admit you're ignorant of the security situation in East Asia and we'll go on. "Who are they defending themselves from". . . what a Chomskyite statement. Believe it or not, a military is in fact necessary even for pacifist regimes like Japan.
Re:US:bombs vs. Japan: environment (Score:3, Interesting)
the U.S. feared that the same thing would happen to Japan that happened to Germany after WWI, that is, that Japan would get strong again and attack.
I'm more of a believer that the US foreign policy folks realized that making Japan artificially weak, in the manner that Germany was treated post WWI with their 'reparations' penalties, would be to repeat a horrible mistake. This would only generate resentment amongst the people, paving the way for 'dynamic leadership' i.e. another fascist/totalitarian government, this time in Japan.
By re-making Japan in their own image, the Americans gained a strong ally instead of creating a bitter foe. Why attack the nation that put you back on your feet? There's a lesson to be learned there.
Re:US:bombs vs. Japan: environment (Score:2)
1,000 Teraflops = One PeatFlop
100 million dollars for another supercomputer just for bragging rights = One bellyflop
Re:US:bombs vs. Japan: environment (Score:2)
While that is true. It is also true that their intention was (is?) to create a petaflop computer ergo the title to the following link:
http://www.research.ibm.com/journal/sj/402/allen.h tml [ibm.com]
Re:US:bombs vs. Japan: environment (Score:2)
Actually, it is/was not a dictated decision.
Quite the contrary, the US was quite eager to arm (West-)Germany and Japan after WWII in order to have strong partners against the Warshaw Pact nations.
Due to their historical background, both nations have a strong pacifistic movement (at last), which actually opposed against building up any armed forces. Probably to appease these movements (and their neighbours), both armies contain something like defense in their name (~jiei~ / -wehr) and in their respective foundations actually forbade military actions outside the nations boundaries.
Both countries are capable to produce nuclear weapons, having nuclear plants and capable scientists, but strongly oppose to the idea of doing so.
(Which might be easier, when you have a strong ally, who assures MAD, in case of an attack.)
Japan has a spending limit of 1% of GNP
(The US spends about 3%, Germany 1.5%)
Re:US:bombs vs. Japan: environment (Score:2, Interesting)
The irony is that both could achieve perfect success with these computers and still be very far from fixing their problem. Perfect understanding of nuclear explosions is only a tiny, tiny piece of the national security equation, and perfect understanding of how typhoons are born, live and die still leaves you a very long way from knowing how to turn one off.
BTW, my daytime job is in aviation. I wish the Japanese all the success in the world at improving weather forecasting--the current state of the art is a disgrace.
Re:US:bombs vs. Japan: environment (Score:2)
Yes but the U.S. Military is investigating on how to generate weather on demand [wired.com]. To own the sky as it were.
Re:US:bombs vs. Japan: environment (Score:5, Insightful)
Are any of the supercomputer projects in the pipeline expected to be faster than this?
More protectionism (Score:5, Interesting)
Re:More protectionism (Score:2, Interesting)
Re:More protectionism (Score:2)
Not really since the dumping claim was dropped after they reached an agreement that made Cray the exclusive US distributer.
Like that would do anything... (Score:3, Insightful)
The WTO is totally powerless, especially against the US. The only thing it provides is a common forum for working these issues out and for establishing a sort of trade best practices. But when you get right down to it, trade disputes are settled as they always have been, either through discussion, or through various embargoes, tariffs, etc. The WTO may add some legitimacy to a particular countries use of some tariffs, etc, but overall it doesn't provide any significant sanctioning ability.
That's the funny thing with all of the world governmental bodies. They have no real power, they mostly just serve as negotiating platforms. The real power continues to be held by individual nations and there's no evidence that they'll be giving up that power anytime soon.
Re:More protectionism (Score:2)
You might be intentionally burying your head in the sand, but it's a fact that Japanese companies will always choose to increase market share at the expense of profitability. Once they put their foreign competitors out of business and get the market share, they really don't know what to do with it (their 1980s economy burst over 10 years ago and they haven't fixed it yet).
*sighs* (Score:5, Funny)
Global Warming Predictions?...or... (Score:2, Funny)
The Model is to follow ... (Score:2, Interesting)
Predictably the model is rumoured to be still 2 years off target yet - so there is the worlds fastest computer sitting idle for the mean time.
Perhaps I could buy some space to run my webpage off it in the mean time
NY times login generator (Score:4, Interesting)
http://www.majcher.com/nytview.html [majcher.com]
Re:NY times login generator (Score:2)
So when does a computer... (Score:5, Insightful)
Re:So when does a computer... (Score:2)
When it scales above the number of processors you can fit on one motherboard.
Multiprocessing systems are communications-bound for most tasks. The communications network becomes more important than the processor very early on.
Put those spare cycles to good use (Score:2, Funny)
Global warming prediction (Score:2, Funny)
I can give you a global warming prediction right away: If when you fire that baby up, the global tempature rises!
If I let my machine be powerd off by night, I have to turn up the temp before I go to bed. And I only got two old P-II CPUs running.
Non-weapons research(in addition to what)? (Score:4, Funny)
This is of course in addition to weapons research of course.
Non Weapon Research? Not according to Dr. Evil (Score:2)
Weather research my butt!
Possibilities. (Score:2, Funny)
1) Make a metal that looks like plastic. Handy for all of those rocket launches.
2) Genetically engineer large reptiles to guard their country from invaders.
3) One word: Gundam.
4) Launch theoretical bombs at ASCI White and see if they can finally win the technology war.
5) Create a fully aware computer program that will help guard us from ourselves.
6) Make a fully synthetic actor that can outact, say, Keanu Reeves. (Oh, sorry, that was the Thunderbirds).
What other possibilities can this thing hold?
What about Google? (Score:2)
Not even close. (Score:2)
Google can NOT do 36.5 TERAFLOPS.
The japanese computer is bigger than the top 10 US supercomputers combined. DO you mean to say google is bigger than that?
ANd btw, this project has been in the works for years, I remember reading about it in some science magazine 3 or 4 years ago, when they started the project.
Pictures here! (Score:5, Interesting)
Isn't this how the nuclear arms race started? (Score:2)
They test one.
Ad infinitum while the world cringes in fear.
It's going to get ugly when Cuba starts hosting Japanese built systems.
[Okay. Lame joke. It sounded better before I typed it, but I'm too attatched to the effort to not post. You're Welcome.]
-Fantastic Lad
Imagine a ... (Score:5, Funny)
Damn, after all these years.... (Score:3, Funny)
Re:Damn, after all these years.... (Score:4, Funny)
New news! (Score:2, Insightful)
Question - why is it that we JUST found out about this? How long did it take to build this giant supercomputer? Companies like IBM talk about what they're building long before they are done. Speaking of which, I guess IBM's Deep Blue is kinda underpowered now, relatively speaking.
One more thing - why all the hub-bub about US export restrictions re: computer power? If Japan already has this much computing power, who wants our "junk" anyway?
I lied - one more thing - does the NSA have penis envy over this? Or is their computer still faster?
Re:New news! (Score:2)
You *JUST* found out about it because you get your news from a narrow band of news sources.
This was in Scientific American, or something similar 3 years ago or so, when the project was starting.
Comutation required for climate/pollution modeling (Score:3, Interesting)
To do a detailed wind modeling, and have a finer resolution, and to do some statistical analysis of different input conditions... suddenly we end up with requirements far more than the current computing power.
We can always come up with a problem that is more complex than we can solve using current computing power. That is a good pursuit.
S
Hey cool. (Score:2)
Nice to see it working now.
Imagine a Beowolf cluster of these?! (Score:2, Funny)
This is very good for lots of reasons (Score:2, Insightful)
After all they don't want just anyone poking around and finding things they shouldn't.
But with non weapons research systems I can see academics from all over the world getting easier access and maybe something interesting can happen.
Re: Possibilities (Score:5, Funny)
Eliza did that several years ago.
This computer is so fucking fast that... (Score:2)
Wargames (Score:2)
"Would you like a nice game of chess?"
very impressive, ... (Score:2, Funny)
Tom's Hardware (Score:4, Funny)
What is undisclosed though (Score:2)
Also keep in mind that several years ago the US govt complained about the French performing nuclear testing under the rubric that they could do it all on a machine. And low and behold only a few weeks ago the DoE 'announced' that they now have the capability to do that, seemingly forgetting that it was previously announced in 1999. So in the intervening 3 years how far do you think they've come.
You know, there are scads of scientists working for the govt who could probably get on the short list for the Nobel if they were allowed to publically publish... and that's basic research. Imagine what applied engineering looks like..
total number of processors is 5120 (Score:5, Informative)
the machine is constructed from 640 nodes, with 8 vector processors per node, and 16GB RAM per node. That totals 5120 processors and 10TB memory.
See http://www.es.jamstec.go.jp/esc/eng/outline/outli
Also of note:
peak performance per processor: 8 GFLOPS
total peak performance: 40 TFLOPS
Hype (Score:4, Informative)
In reality, most of the time, performance is way below peak values, even for the algorithms for which the computer was designed to handle. IBM's pacific blue has a peak TFLOPS value around 3.6TFLOPS...but in reality, its usually around 1.2TFLOPS.
There's no reason to believe this machine will be any different.
Furthermore, the performance of this machine is likely to sink like a rock when its used outside the area it was specially designed for.
In other words, the best supercomputers in the world are still the ones made by starbridge systems, which were bought by NASA (I believe the one NASA bought was called HAL 15, or something like that).
Wow... (Score:2, Funny)
[jamstec.go.jp]
http://www.es.jamstec.go.jp/esc/gallary/images/
"I'm sorry, I can't do that Dave...."
35 teraflops. Wow! (Score:3, Interesting)
SETI operates at 17 teraflops [berkeley.edu], but at a cost of only $500000.
Re:Why so few processors ? (Score:2)
I would imagine that the processors are specifically made for this application, and not some off-the-shelf processor. Also, It much easier to design/build a 5104 processor machine then a 1 million processor machine. Economy of scale doesn't apply here.
Re:Why so few processors ? (Score:4, Informative)
The basic problem of adding more and more processors is keeping all the memory in sync. If you have a process that is running across 50 cpus the machine needs to ensure that if one of them updates a variable that all the others work with the current value. (Ok, it's more complicated than that but I'm not writing a book here)
The solution is to write your system so that the calculations can run as independently as possible. However, at 100 million processors it probably just doesn't fit the problem space.
Re:Why so few processors ? (Score:3, Informative)
The basic problem of adding more and more processors is keeping all the memory in sync.
That's why message passing is typically used instead of some sort of shared memory approach. You eliminate the synchronization problems as well as memory contention. After that, it's just a matter of keeping all the processors busy.
Re:Why so few processors ? (Score:3, Informative)
Wrong. Just plain wrong. Explicit message passing can often reduce communication overhead compared to coherent shared memory, but the synchronization problems are still very much present. You still can't operate on data before it becomes available, regardless of the programming model. Explicit message-passing systems handle synchronization very differently than shared-memory systems, but those problems don't just go away.
Re:Why so few processors ? (Score:2)
So each processor is connected by the front side bus. When you need a value you check each processor to see if it has it before you get it from main memory. Your alternative is to check with the controlling processor instead of each individual processor. Through the years they found that the loss of a processor just for admin tasks wasn't worth it. Now everybody shares admin load and everybody does work.
Because this admin load get's to be too much, they normally divide the machine into subsections. For example, the 64 CPU sun box I use at work is divided into 8 smaller sections. Each of the 8 CPUs are equally close to memory but they are far from the other "sub machine's" memory.
Anyway, I'm kind of rambling here but the general idea is that super computer builders have moved away from that idea in most of the models I've seen. Instead they try to keep the communication requirements low. (AKA, maximizing locality of reference)
If a "real" expert is around who knows of something different I'm all ears.
Re:Non Weapon research?? (Score:5, Informative)
Yeah right !
Uh.. from Chapter II, Section 9 of the Japanese constitution:
"Aspiring sincerely to an international peace based on justice and order, the Japanese people forever renounce war as a sovereign right of the nation and the threat or use of force as means of settling international disputes. 2) In order to accomplish the aim of the preceding paragraph, land, sea, and air forces, as well as other war potential, will never be maintained. The right of belligerency of the state will not be recognized."
The Japanese are only able to maintain a defensive force, not an army, so even if it was weapons research, it would only be for use in self defense.
Re:Non Weapon research?? (Score:3, Interesting)
It seems a largely successful strategy and it might be better if more countries were to consider it.
Re:Non Weapon research?? (Score:2)
So long as "more" is "all". If one country doesn't renounce violence it can just take whatever it wants from the other countries. If several don't they could just split up the world...
Re:Non Weapon research?? (Score:4, Insightful)
Re:Non Weapon research?? (Score:2)
Re:Non Weapon research?? (Score:4, Informative)
Re:Non Weapon research?? (Score:2)
It's probably just a not particularly subtle jab at the US DOE nuclear weapon simulations research which gets done on the big American gov't supercomputers -- in other words, pointing out that they're using their CPU cycles for what theyc consider a better purpose. Japan isn't particularly fond of nuclear weapons at all.
Re:Australian SuperComputer uses (Score:2, Funny)
Re:Australian SuperComputer uses (Score:2, Funny)
Re:Processor number & Beowulf (Score:2)
Ahem, I think you meant to say 3.2 Million processors. Not Trillion unless the math from your Universe is different than the math from my Universe.
Re:Processor number & Beowulf (Score:2)
Oui.
My point is that it's a very _big_ number for beowulf clusters. The biggest one have 8192 processors
Hmmm...I thought that Beowulf referred to separate computers and not processors. So five computers linked together are a Beowulf cluster of five while a single computer with five processors is not a Beowulf cluster of anything but simply a multiprocessor computer, or is this a incorrect interpretation?
As for the number of processors, I'm sure that they had to make up for the lack of processor speed with sheer numbers. Yuck!
Re:Processor number & Beowulf (Score:2)
A cluster is a cluster, there are many different kind. There is no one kind called a Beowulf.
Beowulf was the name of a project at NASA that was building clusters out of cheap computers. So I guess any cluster built out of cheap computers is a Beowulf.
Beyond that, there is no set standard for how a "Beowulf" cluster operates. They all use different librarires, different cabling, etc.Some use PVM. Some just use mosix. Some use other stuff. Etc..
Re:Take that ASCI White!! (Score:2, Funny)
Re:Take that Doppler 10,000 (Score:2, Funny)
I wonder how far in advance this new supercomputer can predict how far John Bolaris [nbc10.com] is going to be off in his predictions again (the poor guy made some completely overhyped predictions about a blizzard last year in Philly area).
Anyhow, hats off, Japan! I'm impressed.
Re:Non-weapon? (Score:4, Funny)
You know you're right!
- scientist: our climate modeling indicates that if we start our weekly barbeque at exactly 6:17pm, a US weapons lab will be destroyed by a powerful tornado in 41 days.
- director: well let's start our barbeque at 6:17pm to see if you're right. Welcome to the 21st century, America! (insert maniacal laughter).
CPU affinity (Score:2)
Pocket? What's the point of playing Quake in your pocket? Or maybe I shouldn't ask...
640*5104==3.2M CPUs... so I can dedicate four CPUs to each pixel on a 1024x768 display, and get reasonable Quake performance without hardware acceleration? (-:
Re:Glad we are doing weapons research (Score:3, Insightful)
That many years ago the US 'forced' them to be pacifist, so now they work on climate research rather than weapons research?
You are one of those poeple who carries a grudge around forever, from your ancestors ancestors, right?
Yes, it is better to simulate a nuclear explosion than set one off. Kind of.
Is it actually cheaper? I don't know how much the computing time costs to simulate one.
The fact remains. THe japanese just built somethign absolutely marvelously huge to study climate, not bombs. And that is noble.
Re:How do you convert code to vector fomat? (Score:2, Informative)
The real trick to Vector code is to work with the memory subsystem and not have it permanently trying to catch up with you as in normal processor style.
You can pretty well put as many floating point units in a modern cpu as you want, the problem is feeding them with data to operate on and storing results. Current microprocessors use multilevel caches to try and keep what it hopes are useful subsets of main memory close to the cpu. Trouble is if you are scanning 100's of GBs of data in a weather model there may be virtually no useful small subsets.
For vector processing you design a system where, before you actually fire off any calcs, you give the memory system a list of the next 64/128/4096(varies) addresses you plan to use. It may take a little while to get the first one but after that they arrive at 1 per clock per memory pipe and, depending on the number of memory pipes you use you can actually drive your floating point units full speed.
Because you want to process streams of memory addresses as a single op (vectors) you spend all your time looking for loops where each iteration can be calculated independent of the next and where the compiler can be sure of that with no ambiguity. That tends to mean no subroutine calls, anything a(i)=f(a(i-1)) is bad but a(i)=f(a(i+1)) is fine and even a(i)=f(a(i-65)) can be OK depending on vector register length. You then get into CIGS (compressed index gather scatter) ops like a(i)=b(c(i)) and you can work with that sometimes etc.
Bottom line, if you don't vectorise high 90%'s of your code the Vector computer is a very expensive room heater. You then need to worry about 99% parallel code+ for multinode architecture but there are similarities between data independence of vectorised loop coefficients and parallel modules.
HTH
Crash
Re:Weather Man, tell me when to launch ICBM (Score:2)
I'm not going to even get into it, but the Constitution you refer to was written by Americans, not the Japanese themselves. The repeal of Article 9 [wsws.org] has been debated for many years, and Japan may well repudiate it in the next few years, and become a "normal" nation with seagoing navy and overseas bases.
Re:Japanese are just smarter and more resourceful (Score:2)
Re:Riiiiight... (Score:2)
Global warming research is weapons research. They're just digging for global warming propaganda, a good part of any war machine.