The UK's Fastest Supercomputer

bmsleight writes "The Guardian has a story on the HECToR, The largest supercomputer in the UK — around five times more powerful than its predecessor, HPCx, which is also at the University of Edinburgh. It measures up well internationally, sitting at 17 in the top500.org list of the most powerful computers in the world."

Comments

UK commitment to science

[BWJones]

I have really been impressed with the level of commitment to science, research and education outside of the US right now and efforts like HECToR only consolidate that impression. While we here in the US have essentially dropped the ball on education and science funding for the past oh, six or seven years, the rest of the world is really stepping up. Of course I have mixed feelings about this as I am a US citizen who works in science and education, but it is also good to see other countries stepping up. For instance, a few months ago, I visited the University of Leicester and was truly impressed with the focus and quality of the research going on in the UK. Their commitment to bioscience funding is something that the US government should be very careful about as we stand to lose some valuable talent overseas if we are not careful...

Re:UK commitment to science

[oojimaflib]

In other news, bioscientists at the university of Leicester have developed a new species of grass. It promises to be at least two times greener than comparable varieties in the US.

Re:UK commitment to science

[Anonymous Coward]

the US have essentially dropped the ball on education and science funding for the past oh, six or seven years, the rest of the world is really stepping up.

what a common shill. if you honestly think this problem is so new it shows that your head has been buried in the sand like the rest of the shills who keep cawing on about the democratic party. you're part of the system that is allowing this type of decline to go on. both big parties are against individual advancement and both want to keep the power firmly in their palms where you'll have to answer to them or be cast aside. have fun. this situation will not improve as long as you keep buying into a system that has been corrupt for decades.

Re:UK commitment to science

[Silver Sloth]

The University of Edinburgh has a long history of being at the cutting edge in computing. I worked there in 1972 at the School of Artificial Intelegence under Donald Michie and Robin Popplestone (well, I washed up the coffee cups!)

Of course, if you ask a Scot, then most of the major technological advances of the 19th century were made north of the border and that proud heritage is alive and well today. Sassenachs may differ.

Re:UK commitment to science

[damburger]

I'm studying physics at Leicester, its a good university. Don't be too pessimistic about the US though, a lot of the research done in our department is in collaboration with the US. I myself am involved with the Leicester cubesat project and we are currently looking to work with the University of Florida.

Re:UK commitment to science

[Anonymous Coward]

For once in a long while it appeared we had a well researched, interesting submission devoid of politics or US bashing.

Re:UK commitment to science

[everphilski]

Yeah, I mean, the US only has 10 of the top 16 supercomputers ahead of the UK.

And you may say "we here in the US have essentially dropped the ball on education and science funding for the past oh, six or seven years", but as a college graduate in engineering from (early in) that time frame, with younger siblings (my youngest is 11 years my junior) interested in education/scientific fields: one brother a pilot with a BS, one becoming a teacher, my sister studying to be a medical doctor and my youngest brother still in high school, but very into science - I'd have to disagree. I could go on and on ...

Re:UK commitment to science

[BWJones]

Ah, but you see I am looking *down* the road a little bit and now where we are currently at. Congratulations to your family on their current or future academic accomplishments, but you have to know that the statements that I made are based on factual conditions of funding from both the NIH and the NSF whose budgets have not even kept up with inflation.

but as a college graduate in engineering from (early in) that time frame,

You might have picked the right field for short term gains.

with younger siblings (my youngest is 11 years my junior) interested in education/scientific fields:

Things may be fixed by the time your siblings are interested, but it will take at least a decade to fix the damage that has been done to science and science funding over the past several years. In the early 90s we spent much effort funding science and education and encouraging students to go into these fields, only to pull the rug out from underneath them when it came time to have them get started becoming independent scientsts. I've been fortunate in terms of funding and worked hard to maintain our position, but many junior (and senior) scientists are very worried about their funding.

one brother a pilot with a BS,

Then both you and he should know what a mess our current domestic airline industry is and unless he is a pilot for Delta, he is not doing nearly as well as he used to before the airlines had to deal with the increased costs of security, delays due to insufficient infrastructure, fuel costs that have tripled, etc...etc...etc...

one becoming a teacher,

God bless them for going into such a low paying career. I briefly attempted teaching junior high school before returning to graduate school when I realized that even as little as a graduate student makes, it was still more than what a teacher makes. If we truly placed a value on our teachers, we would not have the lack of commitment to the profession in terms of requirements for standards and low pay.

my sister studying to be a medical doctor

I am a principal in a medical clinic where we have about a dozen docs, our own MRI and CT scanners and about 100 total employees. On top of that, I teach medical students and am involved in the selection of medical students at my university. I think that I can say with some authority that medicine in this country has changed and not for the better. Even worse, we have not made any progress over the last few years on fixing any of the inherent problems with providing medical services in this country and in fact, have accelerated the damage being done by further limiting our options. Your sister is heading into a profession that is horribly broken in the US and is in need or a dramatic overhaul. Hopefully she can be a part of the solution...

Hey, in fact, we are in such desperate need of physicians if you know a neurologist or a cardiologist that wants to joint our practice, send them my way. If we hire them, I'll cut you a check on the spot for $10,000. I am serious. There are rural places in this country where physicians are simply, almost impossible to find.

and my youngest brother still in high school, but very into science

Cool. As one in science, I would very much like to encourage him. But we need to fix things to enable us to continue to stay a leader.

- I'd have to disagree. I could go on and on ...

Because we live in a (mostly) free country, that of course is your prerogative. But ask anyone in the trenches of science and education and they would have to be honest with you and say how things are. From this scientists/educators perspective, we need to change our approach.

Re:UK commitment to science

[everphilski]

You might have picked the right field for short term gains.

I'm in Aerospace. The industry isn't going anywhere ...

I've been fortunate in terms of funding and worked hard to maintain our position, but many junior (and senior) scientists are very worried about their funding.

I'll grant you, I work in engineering more than the science fields, but I haven't encountered that. In fact the school I attended is looking to hire 5 more professors in the next 5 years, in the Mechanical Engineering department. That's a staff increase of 20%, for a school not known for it's ME department. I think that's an excellent sign.

Then both you and he should know what a mess our current domestic airline industry is and unless he is a pilot for Delta

He's working for a courier service you've never heard of and I can't recall in the midwest. Flying for the big guys has its perks (and is his eventual goal) but the advantage of this job is he comes home to his own house (and fiancee) every night.

I can't really speak to medicine. I'll take your word on it. Other than seeing a specialist a few times a year for chronic migraine, I do my best to stay away from doctors :)

But ask anyone in the trenches of science and education and they would have to be honest with you and say how things are. From this scientists/educators perspective, we need to change our approach.

I won't argue, and I'll take your word on medicine, but at least in my end of the world (aerospace engineering), things are on a definite upswing, and it's not all application there is a lot of research going on as well. I interface with a lot of university professors and PhD's, I have three close relatives who teach at the grade and high school levels, and I just don't get the vibes from any of them that I read about here or other places. Maybe each of us is just blessed being where we are at.

Re:UK commitment to science

[Bloke down the pub]

Blah blah USA blah blah number one number one num be ONE. Blah blah my brother blah.

Whatever. Get back to us when you don't believe that the Earth is six thousand years old and dinosaurs were hunted into extinction by cavemen.

Re:UK commitment to science

[xaxa]

I grew up in/near Leicester, it's a great city :-D. It's interesting to read an account of it by a foreign 'tourist', it inspires me to investigate the city some more.

BTW, it's not really a college town, so it doesn't feel too empty in the holidays (compared with somewhere like Loughborough, anyway).

PS -- you can avoid two-hour cab rides to London by taking the train, 90 minutes each way.

Re:UK commitment to science

[BWJones]

Thanks and Leicester is a great town. We took the cab simply out of convenience for our hosts. Door to door service was a bit more expensive, but ultimately convenient. I wished to have a bit more time to explore some of the Roman ruins and sample some more Indian food which seemed better than it was in London.

Re:UK commitment to science

[Anonymous Coward]

Apparently the locals don't share your enthusiasm for UK science commitment:

http://space.newscientist.com/article/mg19626356.100-science-cuts-may-harm-uks-international-reputation.html

Re:UK commitment to science

[Anonymous Coward]

"I'm in Aerospace. The industry isn't going anywhere ..."

Worst choice of words, ever! (Uh, I hope!)

Re:UK commitment to science

[Anonymous Coward]

PS -- you can avoid two-hour cab rides to London by taking the train, 90 minutes each way.
Actually, it's better than that - Leicester to London St Pancras is 75-80 mins on the twice an hour fast service at xx00 and xx30. 90 minutes is for the stopping services which leave 5 mins after the fast services.

Of course, if they electrified this route and made some minor upgrades it would only be about an hour.

Re:UK commitment to science

[everphilski]

How do my beliefs as a single individual factor into the status of the US at large? And how do you claim to know my beliefs, which by the way, are not as you state?

Re:UK commitment to science

[somersault]

Yep, Functional MRI (fMRI) was developed at the University of Aberdeen, where I studied (sure I was studying Computer Science and not biology or whatever, but it's still pretty cool). Sad thing is how the first poster seems to think that smart people only exist in the US, either that or he thinks that having zillions of dollars is all that it takes for advancement (sure it helps a *lot*, throwing money at problems always tends to help, but good education is also very important)

First application

[mrami]

Figure out how Brits can stand Marmite.

17th isn't good enough

[damburger]

The UK GDP is 5th in the world (nominal) or 6th in the world (purchasing power parity). If our best supercomputer is coming in at 17th, we aren't spending enough on research.

Not to belittle this project, of course, building the worlds 17th fastest supercomputer is an achievement in anyone's book - but it is a sign of where the UK government is weak.

Re:17th isn't good enough

[AvitarX]

Wouldn't it make sense to compare total computing power to GDP, or at the least total scientific computing power?

I would doubt for example you would have the same complaint if the UK had the 17-100 spots on the list.

It could very well be that the UK is spending a lot more on research, but does not like to spend it on large super computersm or even spends it partnering with facilities in other countries.

Re:17th isn't good enough

[Ed Avis]

I don't think research is a pissing contest to see who has the most powerful computer.

The metric for 'most powerful' also seems flawed. If you just count operations per second, then a large enough cluster of Linux PCs will appear 'more powerful' than any supercomputer, even if they are connected by UUCP over 2400 baud modems. Yet the supercomputer is much faster at most difficult computational tasks because it has faster connections between the nodes. The Linux cluster would only outperform it for drawing a large picture of the Mandelbrot set or other 'embarrassingly parallel' problems. This isn't an academic distinction; even gigabit Ethernet is much slower (higher latency) than the links used in a real supercomputer.

top500.org do categorize each supercomputer as 'cluster' or whatever but I think their ranking is just on raw operations per second.

Re:17th isn't good enough

[damburger]

I'm working on the assumption that the computing power of the newest supercomputers dwarves that which has gone before, and that's not an unreasonable assumption. Do you have any numbers for total computing power by country?

Re:17th isn't good enough

[Anonymous Coward]

The UK GDP is 5th in the world (nominal) or 6th in the world (purchasing power parity). If our best supercomputer is coming in at 17th, we aren't spending enough on research.

I'm not convinced this is a logical conculsion. 11 of the faster computers are in the US, which is a much bigger, richer nation than the UK and very much a special case. So we're 6th out of the remaining countries, which seems fairly reasonable. Anyhow, a more meaningful measure of whether the UK is punching its weight might be something like 'total supercomputer capacity per unit of GDP' or something like that.

Re:17th isn't good enough

[damburger]

Thats a fair point actually. If the UK is the sixth country on the list, not sixth supercomputer, we are doing OK. I'd still like to see a comparison of total computing power though.

Re:17th isn't good enough

[AvitarX]

Well if you want me to use information that's available (I hardly think that's in the spirit od /.), I would use R&D total numbers to determine the amount spent on R&D, and not limit myself to computing (after all not all research is super computing).

If I do such, a quick googling finds the UK is 4th worldwide (http://www.aaas.org/spp/rd/guiintl.htm). here (this is a google cache link, to view a PDF as HTML) is a table showing it just above China as a percentage, but about 3/4 of the highest percentage (korea) and a little less behind France, Germany,US and the Japan. And if the 2 sites are using similar methods we see the UK just above the EU-27 as a whole.

The UK is certainly not the top that one may hope, but it does appear to be above average (4th place in raw number).

Re:17th isn't good enough

[Comatose51]

Well the obvious answer for the UK then is to produce less goods until they are in fact 17th in GDP.

Re:17th isn't good enough

[Leebert]

The UK GDP is 5th in the world (nominal) or 6th in the world (purchasing power parity). If our best supercomputer is coming in at 17th, we aren't spending enough on research.

Nonsense. You build the computer that is as large as necessary to get the job done. I, for one, am sick of the HPC "mine is bigger" envy. You have N science to do, which requires X amount of computational resources. Buy something close to X. If that means you're 17 on the top500, so be it.

Disclaimer: I work in an HPC shop (which has dropped quite a bit down in the top500 over the last few years), so I'm probably grouchy and jaded about the subject.

Re:17th isn't good enough

[photomonkey]

I'm out on a limb here, but I'm betting the computer was financed by private donation and public funds/subsidy.

In which case, the university has to ask itself what it needs. How much 'worse' is the 17th fastest box from the 5th or 1st fastest? Does it meet the needs of the university over the length of its amortization?

Re:17th isn't good enough

[iLoveYoyo]

stupid arrogant Americans, waiting for death!

Re:17th isn't good enough

[some damn guy]

Good point, but the other big hole is what "faster" really means. 1-16 are faster at running linpack, but that's about all you can tell. The linpack benchmark isn't horrible, and it's a long-time standard which counts for something, but the truth is it's really out of date. You wouldn't care how many fps you got on Doom II if you bought a brand new computer, would you? You would be needing to move a lot more data with a modern game, and so memory bandwidth, specialized hardware etc would matter a lot more, but if you just cared about doom performance you could skimp on all that and just go for raw clock speed. Not an exact comparison. mind you, but you get the idea. There's a lot of benchmarks you could run that wouldn't show any improvement from a $600 graphics card over a $10 one but obviously there's still a market for them, it just depends on the job.

This isn't a commodity cluster, every opteron is connected with it's neighbors in 3 dimensions via a custom router that plugs right into the hypertransport bus. There's no southbridge/pci card or something like that in the way. There's just a ton more node-to-node bandwidth and a lot less latency. You won't see the difference on linpack or seti@home, but you will on real-world scientific simulations like climate modeling, especially when you talk about 100s or thousands of processors working on the same job. Seti@home doesn't slow to a crawl because Joe in Toledo discovers bittorrent, but these types of codes are like that.

Trust me, this is a world-class machine for science. For the money they spent, they could be a lot higher up, but you don't bring a dragster to the Indy 500.

Second Application

[skoaldipper]

Figure out how to keep Hector from a murderous rampage on Saturn 3. Run, Farrah, run!

17th isn't bad

[magarity]

It measures up well internationally, sitting at 17
 
The British don't mind being at any number as long as the best French *whatever* is lower ranked - 19 in the case of the latest supercomputer list. Although they might be a little out of sorts that Spain is above them at 13.
 
Note: if you are British or have any British friends, the above is 'funny' or 'insightful', not 'flamebait' or 'troll'.

Re:17th isn't bad

[funkatron]

Note: if you are British or have any British friends, the above is 'funny' or 'insightful', not 'flamebait' or 'troll'.

Anyone that needs telling is clearly not British.

Re:17th isn't bad

[Bearhouse]

As an Englishman, I'll go for funny - far too many troll & flamebait mods here IMHO.

But I'd say that recent rhetoric against France has been far more virulent from across the pond.

BTW, the 'old enemy' was traditionally the Catholic alliance of France AND Scotland. Although Spain and Germany have featured heavily too..

Anyway, we'll see how you gentlemen react when China and/or India builds a bigger one than yours. He who laughs last...

Re:17th isn't bad

[Anonymous Coward]

BTW, the 'old enemy' was traditionally the Catholic alliance of France AND Scotland. Although Spain and Germany have featured heavily too..

How politically correct of you to leave out the Irish...

Re:17th isn't bad

[Anonymous Coward]

The Irish gave the world Guinness, all is forgiven.

Re:17th isn't bad

[Dogtanian]

The British don't mind being at any number as long as the best French *whatever* is lower ranked - 19 in the case of the latest supercomputer list. I'd say us Scots feel the same way about having the UK's fastest computer rather than the Sassenachs ;-)

Re:17th isn't bad

[Bearhouse]

Not PC..that was the original usage. You also forgot the Welsh. Nowadays, of course, even the Australians also use the term 'the old enemy' (incorrectly) to refer to the English.

Re:17th isn't bad

[caluml]

France: Always below Britain, in all senses.

I keed, I keed, my French chooms.

Re:17th isn't bad

[Anonymous Coward]

As another Englishman - I will go for 'funny but true' and hats off to India (number 4 and rising?)

Re:17th isn't bad

[owlnation]

You need to look up "Sassenachs" in a dictionary. And then note carefully the geographical location of the city of Edinburgh.

You might, too, want to note that the University of Edinburgh has a very low percentage of Scots-born students and staff.

It wouldn't hurt you to also note that "Scotland" is a figment of your imagination. It's a collective delusion without any legal status or basis in current fact. While it was 400 years ago, today it's not a country, not a nation, not a state, -- merely a convenient way of defining a geographical area where certain local laws apply. It has very little autonomy - and far less than any German, Spanish or US state does.

Do you seriously think that Glasgow, Edinburgh, Perth or Dundee have ANYTHING in common? Glasgow has far more in common with Liverpool, Manchester or Newcastle than anywhere in "Scotland". The parliament is just a new way of scamming more local government funds without any true representation. In this, it replaces the Regional Councils and takes their scams up one notch higher.

Sorry, but despite being born in Edinburgh, I'm seriously tired of small-minded, chip-on-the-shoulder Scottish-Nazism.

Re:17th isn't bad

[MarsDefenseMinister]

The only number that matters in that regard is 1588.

Re:17th isn't bad

[Anonymous Coward]

So, the UK might be the greatest place to live and die. You still have to put up with your weather.

I'll take France anytime, thank you. :p

Re:17th isn't bad

[Anonymous Coward]

The French are 19th?

Phew. They still have better internet than us, though... Grumble grumble...

Re:17th isn't bad

[PatrickThomson]

I'm seriously tired of small-minded, chip-on-the-shoulder Scottish-Nazism.

Don't you mean "deep-fried-mars-bar-on-the-shoulder Scottish-Nazism"?

Re:17th isn't bad

[dartmongrel]

Regardless, good whisky.

Re:17th isn't bad

[dartmongrel]

Except that we have better food (goes without saying) and our women are hotter. Don't like France? keep your sunburt hide off our beaches then, whot?

Re:17th isn't bad

[Dogtanian]

My God.

I thought my comment was *very* obviously a tongue-in-cheek response (in the spirit of the post being replied to), making fun of the rivalry in a lighthearted manner. I was faintly worried that someone would take it in the wrong spirit, so I left the winking smiley on.

And there's *still* someone out there who takes it the wrong way. *And* reads way more into it than there was *and* uses it as an excuse (intentionally or otherwise) to air the bees in *their* bonnet!

You need to look up "Sassenachs" in a dictionary. And then note carefully the geographical location of the city of Edinburgh. Funny, I looked it up in two dictionaries, and they both claimed that it referred to the English.

Yes, I'm well aware that its origins were probably different, and that some may argue that it still holds those meanings, but frankly, I don't intend getting into a nitpicking discussion with the type of person who overanalyses, misses the humour in and misses the point of a throwaway joke.

You might, too, want to note that the University of Edinburgh has a very low percentage of Scots-born students and staff. To be blunt, had I been serious, I wouldn't have claimed that this that specifically a "Scottish" achievement- I believe that it was probably done in conjunction with lots of people, including those at other universities.

I'd half-expected someone to make another joshing response (in the same spirit as mine) along those lines- would have been quite funny then. More than can be said for you.

It wouldn't hurt you to also note that "Scotland" is a figment of your imagination. [..snip..] Do you seriously think that Glasgow, Edinburgh, Perth or Dundee have ANYTHING in common? [..snip..] The parliament is just a new way of scamming more local government funds [etc] Seriously, did you just use my one-line joshing as an excuse to launch this rant?

Sorry, but despite being born in Edinburgh, I couldn't give a damn where you popped out of your Mum's womb. Regardless of who you are, what you consider your nationality to be, whatever, you're still a dick.

I'm seriously tired of small-minded, chip-on-the-shoulder Scottish-Nazism. So, from a one-line comment that any reasonable person would (at worst) take in the spirit of friendly rivalry and (at best) see as a tongue-in-check acknowledgement/parody of that rivalry- a one line comment that includes a winking smiley for fuck's sake(!), you manage to read all that into it and end up trying to associate me with Nazism?

You humourless, odious prick.

Your reply says far more about you than it ever will about me.

Chip-on-the-shoulder? You blatantly have a chip on *your* shoulder if you managed to get all that ranting out of a mildly funny joke.

Re:17th isn't bad

[Anonymous Coward]

And meanwhile, the Germans sit at #2... I wonder what the Britons think about THAT.

But does it know...

[Billosaur]

...the answer to Life, The Universe, and Everything?

Re:But does it know...

[DMoylan]

are you thinking of hactar, designer of the ultimate weapon?

http://hhgproject.org/entries/hactar.html

Re:But does it know...

[Anonymous Coward]

42

Re:But does it know...

[eno2001]

Redundant? Come on you stupid mods! Obvious? Sure. Redundant? I think not. Personally I would have modded it +1 Funny.

Re:But does it know...

[Billosaur]

It's ok. I find it humorous that it was modded both "Redundant" and "Interesting"...

Location, location, location

[Anonymous Coward]

The previous generation one is *not* located in Edinburgh, it is at Daresbury Laboratory. Get your facts straight, Guardian! In both cases, Edinburgh is a *partner* in the HPC running - it does not belong to them.

Cray XT4 Supercomputer

[icehawk55]

The article really didn't say much about HECToR itself. It's a 60 cabinet Cray XT4 system that currently has over 5500 AMD dual core processors. We'll be upgrading it in stages over the next couple of years to over 250 Teraflops. Including some cabinets of the new Black Widow Vector product, now called the Cray X2 system. The Cray team, myself part of it, is actually a multinational effort. I'm a US citizen who is headed over to maintain the system, we have a Brit on the team and the third is also from outside the UK. It's an interesting situation. The biggest UK system, being maintained by two expats and a local. (-: ice_hawk55

Re:Cray XT4 Supercomputer

[Bazman]

...and it's probably being used by a bunch of French, Danish, Chinese and Portuguese. That seems to make up most of our maths department academics these days. Sometimes I'm the only English person at coffee breaktime. Luckily everyone in the world speaks really good English these days.

  Not that this is a bad thing, nor is it a one-way thing (one of my English colleagues is off to a job in Chicago next week), but it illustrates that so much academic work at the top level is multi-national.

  I was at a meeting last year about discussing the successor to HECTor...

Re:Cray XT4 Supercomputer

[icehawk55]

The follow on system started off being called Son of HECToR but politics has changed the name to Child of HECToR. Still up in the air as to what it will be. But, one needs to plan way in advance when building these animals.

Re:Cray XT4 Supercomputer

[Bazman]

Yes, I remember people in the meeting kept calling it 'Son Of' but had to be corrected! Oh how we laughed at political correctness gone mad!

But is the machine room it lives in called Hector's House? Its instant nostalgia for any Brit kid in his or her late 30s/early 40s I reckon.

Re:Cray XT4 Supercomputer

[icehawk55]

That's the first I have ever heard of Hector's House. I love it. I'll be passing on the link to the rest of the team involved. Thanks for the heads up. (-:

Re:Cray XT4 Supercomputer

[Dave Fiddes]

Welcome to Scotland! I work about 200 yds from HECToR's home. It's a nice part of the world to work just next to the Pentland Hills outside Edinburgh. Quite a trek to get into Edinburgh itself though.

I guess that since the demise of the INMOS Transputer the UK hasn't had much to do with the design of supercomputers...though the architecture of the XT4 does seem to bear some similarity to the Transputer systems of the late 80s/early 90s.

Re:Cray XT4 Supercomputer

[icehawk55]

Don't write off the ACF too much. They have a couple of interesting systems in house, not just HECToR. HECToR just gets the attention because of how big it is and how much it cost.

I spent two weeks over in Scotland the first part of December looking for a house. It's a beautiful area. We'll be living in Newtongrange it seems. Only an 8 mile commute to work. The family is going to love it there.

I can't wait to get there and show them the area. We'll be in country around the 15th of January or so.

Nuclear stockpiles

[chromeshadow]

Does anyone know why increasingly powerful supercomputers are needed to ensure the safety of nuclear weapon stockpiles? Given that these are existing weapons which are (presumably) just sitting around in silos?

Re:Nuclear stockpiles

[icehawk55]

The people who truly understood how our Nuclear stockpile worked are all older and retiring. For years and years it's not been a field of study that was popular with Phd students. Now we're in the situation where the people who know what's going on are retiring and there are not enough new folks coming down the line that understand what's going on. Add to that the fact that you can no longer actually set one of these beasts off and they have been sitting idle for decades. What's the state of the current stockpile? The only way to predict how these decades old weapons will react is to simulate them with a supercomputer as best you can.

Re:Nuclear stockpiles

[canuck57]

Does anyone know why increasingly powerful supercomputers are needed to ensure the safety of nuclear weapon stockpiles? Given that these are existing weapons which are (presumably) just sitting around in silos?

If the politicians don't like the results. They buy a faster computer and run it again until the get the results they like.

fastest decline of a super power in history

[Anonymous Coward]

8 years.

Re:fastest decline of a super power in history

[Anonymous Coward]

Nah, it only took superman a few minutes to lose his powers in Superman 2.

Top500

[prakslash]

Doesn't this Top500 contest boil down to a matter of who has more money than the other?

I mean, at this stage, there isn't any real innovation in interconnect or processor or memory technology. It is mostly a matter of who has the money to buy thousands of these chips, cobble them together and supply enough money to keep the whole thing running.

If University of Edinburgh had thrice the money, they could cobble three Hectors together and then they would have had a system at least twice as powerful or may be only 50% more powerful (Whatvever the power gain is). Then they would end up higher on the list.

May be there should be some kind of constraints built in within the Top500 to encourage actual innovation as opposed to measuring the financial resources of an institution or a country.

Re:Top500

[UKRevenant]

The top 500 is not a very good measure of performance of supercomputers. But it is probably the only simple way of comparing them.

When you order a super computer you do not (or at least never used to) worry about benchmark performance, you were concerned about getting something to solve a problem. The two are not unlinked, just some design criteria do not translate well into generic benchmark performance.

That said, of course being a university the supercomputer may be intended to be tasked with many different problems and for that reason a more generalised configuration may be specified. I have been involved in several projects where very powerful setups have been purchased for a single task and once that task is completed the entire setup is disposed of (broken up and sold off) instead of being re-tasked, which I always felt was a large waste of money.

As for where the UK stands internationally, it would be nice to see a few more supercomputers on the list and certainly higher up, but traditionally the UK has been very good at creating new technologies and creative solutions on a low budget. I hope that the constant tinkering with our education system does not bring this wonderful tradition to an end.

Oh, and a happy new year.

Re:Top500

[afidel]

Huh? One of the best ways to get onto the list is to figure out a new and innovative interconnect topology. Remember System X from Virginia Tech, it was in the top 10 on the list when implemented and cost a fifth of the next cheapest system. It's not all about money unless you are gunning for #1. Virginia Tech is also the creators of an interesting alternative list, the Green 500, a list of the most power efficient supercomputers.

Re:Top500

[flaming-opus]

True: The top500 is mostly a contest of who can buy/power more processors. That doesn't mean there is no innovation going on, it simply doesn't matter for placement in the top 500. Most savy HPC customers know that linpack performance isn't all that important, and they do use other metrics to select their HPC solutions.

Infinaband is a much better interconnect than ethernet, and it's made it into a lot of the top500 systems. IBM's Blue-Gene and Cray's seastar networks are even better and they're represented too. Even within infinaband, we're seeing more DDR products, and 4x cards in compute nodes. Switches are becoming bigger, and the switch links are wider, and some are even going Quad-data-rate. Sun put together a very impressive fat-tree-in-a-box for the tacc system. The next generation Bluegene and cray networks are just around the corner, and NEC dramatically improved the IXS in the SX series. Interconnects have been changing quite a bit in the last couple years, and look like the trend will continue.

Memory is following the curve set by the server/desktop world, but processors have been showing some real innovation. Dual and quad core CPU's aren't the answer to every problem, but they offer a real improvement for some problems, with a low cost/power impact. Cray is working on the next generation of vector and multithreaded processors. NEC just put out a 100Gflop vector processor. Power6 looks like an amazingly fast superscalar cpu. Blue gene and sicortex use low-power processors in innovative ways. Cell, clearspead, and GPU coprocessors are beginning to be deployed in very limited ways; with the promise of a lot more to come.

I hate to be an optimist when cynicism is in vogue, but I see a lot of innovation in the last few years, and a lot on the horizon. How effective some of that innovation is, and how cost competitive, I do not know. It's pretty cool though.

HECTOP is Russian for....

[Anonymous Coward]

HECTOP is Russian for NESTOR.

(can't enter cyrillic chars in here)

Congrats to the Brits

[Anonymous Coward]

It's good to see the Brits develop a supercomputer. Very impressive.

They finally figured out a way to make a computer leak oil!

Re:Congrats to the Brits

[digitig]

But it's a Cray, so it's American, right?
So it won't leak oil. But for all its power, it will perform like a family sedan.

Because ..

[OzPeter]

You can't just take one out of the silo and set it off to see if it goes Bang! [1]

So they are stuck with taking a model of a bomb out of a virtual silo and seeing if it goes Bang! virtually [2]

[1] Your results may vary with the age, size and design of the weapon

[2] Your results *will* vary with the quality of the model, which is related to how fast you can run it.

Except ..

[OzPeter]

The name is HectoR

So you need to transliterate the name with that damn awful reversal of the western 'R' glyph acting as a replacement the russian '' glyph.

So that would give you a russian transliteration of which is phonetically closer to 'nyes-to-ya'

Of course I am not a native russian speaker .. so what would I know anyway?

Darn !

[da5idnetlimit.com]

And here I was hoping that the latest TOR node was an HPC ranked 17 in the "biggerz clusterz" list...

One can dream...

HECToR... Really?

[FesterDaFelcher]

With literally thousands of supercomputers in use, and thousands more being built, isn't it time to stop trying to name them cute little anagrammed names?

Re:HECToR... Really?

[Oktober Sunset]

actually, if you look at the top 20, all the US ones are called thing like Red Storm, or Blue Gene or Jaguar or Thunderbird or other silly macho names, HECToR defiantly has the cutest name, especially with the cutesy lower case o. Awwwwww, so sweet, the next cutest name is at number 32, Queen Bee then there isn't another cute name until number 65, Big Ben.

Re:HECToR... Really?

[Watson Ladd]

If you think Hector is cute name, you need to read some more Homer. He kills Patroclus, leading to the reentry of Achilles into the fighting.

damn Slashdot

[OzPeter]

I hit submit by accident and *then* realised what the OP meant by not being able to display cyrillic chars.

top500 uses linpack to measure performance.

[flaming-opus]

Your example of the 2400 baud modems for a linux cluster isn't completely accurate, as linpack does a little bit of communication, though the point is well taken. The top500 list only uses linpack to measure performance, and linpack represents a very easy problem to solve. Essentially, the top500 list is a list of which machines do a really good job of solving a trivially difficult problem. The hpcc benchmarks (http://icl.cs.utk.edu/hpcc/) are a lot more interesting; though, even these need to be read with some caution.

Ranking supercomputers is a really hard problem. Each application has different needs for communication latency, bandwidth, programming model, cache size, memory bandwidth, and computational throughput. Then you have to ask: how much optimization can I do to the benchmark? Am I going to be able to do the same amount of optimization for each of my applications? How easy is it to extract this performance? The guys writing the software for these things are usually professors or post-docs in the hard sciences, not in supercomputing.

HPCx is not at the Uni of Edinburgh

[Anonymous Coward]

HPCx ain't at the University of Edinburgh. It's partly funded by them but is actually housed at the Daresbury Laboratory site of the Science and Technology Facilities Council, and is partly looked after by the Computational Science and Engineering department at said Lab. Once upon a time, HPCx was in the top ten machines in the world. It's since been overtaken by many commodity cluster architectures and most recently Blue Gene of course.

Gallery link

[Anonymous Coward]

If you want to see the pictures, here they are.

Re:Gallery link

[kramulous]

I'd prefer one of those time lapse videos with Benny Hill themed music.

Sweden at no 5

[Anonymous Coward]

Does anyone have any more information on the unnamed Swedish government agency that has the fifth biggest supercomputer is?

As a swede, I'd kind of like to know. I mean, please tell me it's not the tax authorities.

Perhaps this will help...

[OceanKiwi]

...in making more accurate weather predictions, meaning that Brits can not only talk about the weather today (their favourite subject) but how pants it is going to be tomorrow.

Poor comparison

[MGDruss]

From TFA: HECToR's memory is also impressive. It is 3,200 times larger than that of a top-of-the-range iPod 160GB

Hmm ... so a £113M computer has got 3,200 times the memory of a £230 iPod. Let's see how that stacks up:

For the iPod that's £1.44 / GB.

HECToR comes in at £220.70 / GB ...

Re:Poor comparison

[icehawk55]

160 GB Hard Drive vs RAM. Not a good comparison. They didn't mention the SAN storage we have, the tape backups, etc etc. I know. I know. I'm being picky.

But...

[caluml]

Does it run Linux?

Gaaad, I feel so dirty now, I disgust myself.

Re:But...

[Maller]

Yes. Pretty much everywhere, management workstation, login and I/O nodes, blade controllers, cabinet controllers, and even the compute nodes now.

Sweden's got #5

[richie2000]

Except it's not doing research. It's eavesdropping on all electronic communication passing our borders. Welcome to 1984, say hi to Big Brother.

Re:Sweden's got #5

[Iloinen Lohikrme]

Or it's doing calculations and optimizations about folkhemmet: how much money goes there and from where and why. That job does indeed need super computer :) On the more serious note, do you have any link to the site in question? I tried to quickly search for it from Google, but couldn't locate anything.

Thought are you sure that it isn't just used in something else? At-least here in Finland it has been spoken that our government has a software that they use to simulate what effects their financial and policy decisions would have on Finland, before making the final decision about the question in hand. Having a super computer to do modelling would be more than ideal for that.

Re:Sweden's got #5

[Anonymous Coward]

The Swedish computer is used by the "top-secret" government spooks FRA, which basically has same functions as USA:s NSA.

I think I have to start using a _real_ strong crypto now...

Re:Sweden's got #5

[richie2000]

Thought are you sure that it isn't just used in something else? I wish. It's been delivered to Försvarets RadioAnstalt (FRA) to assist in dechipering and gathering keywords for a upcoming law that would let them eavesdrop on all electronic communication corssing our borders in cables. Historically, they have conducted radio sigint against the Soviets, but after the wall went down, they have been looking for a new job.

Grammar Nazi Karate Chop!

[mooingyak]

I'm working on the assumption that the computing power of the newest supercomputers dwarves that which has gone before...

While both 'dwarfs' and 'dwarves' are commonly used pluralize the noun dwarf, 'dwarves' doesn't quite work for the verb form.

For example, "Even though he's only five feet tall, Bobby dwarfs the dwarves."

and FYI my firefox spell checker doesn't like 'dwarves'. It also doesn't like 'firefox' in lowercase.

And while I'm at it, let me apologize for a mostly useless contribution to the discussion.

Re:Grammar Nazi Karate Chop!

[lattyware]

How was that a contribution to the discussion?

Re:Grammar Nazi Karate Chop!

[mooingyak]

It evoked a response.

Obligatory....

[Anonymous Coward]

But does it run linux... Yes it does!

Damn you

[maroberts]

I came into this article to start writing "Hector's House" comments only to find your mucky fingers had been there, done that and got the t-shirt.

Not hardly

[Technopaladin]

It alows enterprising young folks the ability to create programs like Achi11es(to take down HECToR)
Or A Zeus cluster of HECToRs
an AJAX is a program that can run on it all day and accomplish nothing.

Re:Not hardly

[Watson Ladd]

Then after PRIAM beats ACHILLES, and ULYSSES uses a clever trick to reclaim the top spot, a few computer designs survive the implosion of the TROJAN collaberation: AENEAS, which along with DIDO claims the top for a while (until DIDO suffers a tragic meltdown due to a malfunctioning cooling fan when AENEAS leaves), and then ROMULUS manages to take the top spot. Eventually we get to three name computers: GAIUS JULLIUS CAESAR being the first. Anyway, this has been an epic series of puns.

2008 - Apologies to Stanley Kubric

[Anonymous Coward]

(Note - must be read in an educated english accent)

Open the pod bay doors HectoR.
I am terribly sorry Dave, but I am afraid I cannot do as you request

Anonymous because I'm lazy

Re:2008 - Apologies to Stanley Kubric

[Anonymous Coward]

The accent, it does nothing!

The Spanish is More Powerful?

[JamesRose]

That's what they said about the armada too :P

Mod Points

[Anonymous Coward]

Imagine a beowulf cluster of these...!

Watercooling

[Tristanjh]

I'm sure if they tried a little overclocking they could get to 16.

The arrangement

[bob.appleyard]

The CPUs are arranged in a Torus shape, according to here. I've seen a lot of these parallel computers with this shape. I can't think of how to make Google tell me this, so perhaps someone here could. What is it about the torus that makes it a good shape for this situation? Have other shapes been tried?

I have the feeling that an arrangement where the connectivity of vertices (CPUs) was distributed according to a power law (i.e. a few vertices with lots of edges, most with not many at all) would minimize the distance between any vertex. I don't think a torus gives you that. Maybe I'm looking at it the wrong way though.

Re:The arrangement

[icehawk55]

A torus shape gives you the easiest way to get a short point to point communications path. It's better than a fat tree or a straight mesh type topology.

The Cray XT3 and Xt4 systems us a X Y Z physical connection. So, X is along the rows and modules within a cabinet (width), Y is vertical within a Cabinet (height), and Z is between the rows (depth).

This works fairly well from a maintenance AND a performance view. You can get some other more esoteric structures built, but they have trade offs in performance vs maintainability. This one works pretty well for these systems

Re:The arrangement

[Anonymous Coward]

The best short answer to your question is that the network is based upon a router ASIC that has six external ports. You could build a binary hypercube with such a router but it would scale up only to 2**6=64 nodes. So a 3-D torus is about the highest-dimensionality regular direct network you can build with a degree-6 router that scales up to a system of this magnitude. Each of the six links is one of the two
directions on one of the three dimensions: +X, -X, +Y, -Y, +Z, -Z.

All else being equal, you want more dimensions of smaller radix, since higher dimensionality means fewer mean hops, and fewer mean hops means more global bandwidth. (Why? Each hop consumes a share of the network throughput bandwidth on a router, and that's a fixed multiple of its local injection bandwidth. Oversubscribe that and you start to congest.) Fewer hops also means less latency, of course, but for these applications that's effectively hidden and not as much of a concern. Suppose you had a router of degree 12, say. You could build a binary 12-cube with 4,096 nodes out of them and it would have a mean hop count of 6, or you could build a 16x16x16 torus with the same number of nodes (using double-wide links) and have a mean hop count of 12.

Re:The arrangement

[Anonymous Coward]

A 14x14x28 box (5488 nodes) with tiled wraparound would have a mean hop count of 10.5. Why isn't that used?

Re:The arrangement

[Anonymous Coward]

An even-radix toroidal dimension has a mean hop count of k/4.

So a 14x14x28 torus has mean hops of 3.5+3.5+7 = 14.

Vista loads

[Bobb Sledd]

Ah, got something that can finally run Vista, did you?

first of the gang to die

[Cally]

For those wondering about the "firstofthegangtodie" tag, it's a reference to this song... can't believe that's a popular tag, though, are the displayed values not the most popular tags? *puzzled

UK Builds Upon Auto Industry Experience

[PPH]

They found a way to make a CPU leak oil.

I've used a Hektor

[Wowsers]

I have to say, it really is a super-computer, great for learning on.

http://www.old-computers.com/museum/computer.asp?c=602
(Link might not work so look at google cached copy from following URL)
http://216.239.59.104/search?q=cache:HkrZrUYOXy8J:www.old-computers.com/museum/computer.asp%3Fc%3D602+old+computers+hektor&hl=en&ct=clnk&cd=1

Okay, not the EXACT same machine, but they sound the same.

Other Hektor

[morcheeba]

I thought it was this Hektor.

WOW!!!!!!!!1!!!!!!!111!11

[Anonymous Coward]

Finally a story about SuperCompters on /. that's really about a SuperComputer! But then again it was posted by CmdrTaco and not Zonk or KDawson....

We now go back to are regularly scheduled program, where we call any computer that has more then 1 core a Super Computer....

Sweet!!

[Impy the Impiuos Imp]

> The UK's Fastest Supercomputer
>
> HECToR, The largest supercomputer in the UK...It measures up well internationally,
> sitting at 17 in the top500.org list of the most powerful computers in the world.

Stewie: So, how's tha novel you've been working on? Coming along fine? Getting those chapters down on paper? Got a progagonist who overcomes long odds, maybe learns something by the end?

When will...

[Deliveranc3]

Canada be #1?

D-Wave.

Not true.

[Xest]

The UK has plenty of partnerships with foreign research centres and as such doesn't necessarily need it's own supercomputing power. Where the UK is week in supercomputing it may be strong elsewhere and can hence trade resources with other nations (mostly the US) that do have more computing power available.

Regardless however, out of the systems above the UK's 10 are situated in the US and 2 in Germany, that means only 6 nations actually have faster supercomputers than the UK anyway so using your reference to the UK's ranking in terms of GDP at 7th place on the fastest supercomputing list the UK is still doing quite well.

Finally, I'm not sure that this list pays any attention to distributed computing, so whilst the UK may not have powerful supercomputers, they may well have extremely powerful distributed processing systems instead.

Just because a nation is at x position in terms of a certain ranking doesn't mean it has to be at x position for all other rankings. It's unrealistic to expect that the UK should be at number 17th in all possible areas. Things also change rapidly, it wasn't so long ago that the UK was hands down the world leader for robotics for example, but now that's an industry that the UK has long lost to the likes of Japan, that's not to say the UK is not doing well in other areas instead such as the financial and medical industries in which the UK is still a world leader.

The most important point to take away is that the UK is obviously doing something right to be so high in the rankings for GDP and purchasing power parity so it's questionable if more scientific research is needed in the grand scheme of things. I'll admittedly however agree with you that personally I'd also love for the UK to be responsible for many more pioneering scientific breakthroughs than it currently is.

Hector...

[A New Normalcy]

verb [ trans. ] talk to (someone) in a bullying way. Not cute. Curt, maybe.

Correction...

[A New Normalcy]

...to the caption of the photo of the billowing cables (or are they delay lines?) Cray is in Chippewa Falls WN, not MN. Just one letter. Tuttle becomes Buttle. Sic transit gloria mundi.

17th is deserved

[G3ckoG33k]

Looking at some US statistics of R&D expenditures on a global scale it seems UK is on 18th place, calculated _per capita_.
(http://nsf.gov/statistics/seind06/c4/tt04-13.htm) That would make the 17th rank well deserved. ;)

Israel is number one, followed by Sweden, Finland, Japan, Iceland, and the US.

Of course the absolute magnitude is of importance here. That allows the UK to engage in research never attainable by e.g. the Icelandic however high their per capita is. With the US in at a sixth place (per capita) and with the largest economy in place, we can safely assume they can do any research they wish. And buy any computer there is. The www.top500.org is a testimony of that. http://www.top500.org/overtime/list/30/countries thus also recognizes the UK economy, and gives it a rank 4! So, the original grand grand parent should not complain. The UK is fine.

Don't feed the Troll

[Bearhouse]

"I don't intend getting into a nitpicking discussion"

But you did!

As a Englishman with many Scots friends I had no problem with your orginal post, which was clearly a 'funny'.