Slashdot Mirror
Landmark Calculation Clears the Way To Answering How Matter Is Formed
First time accepted submitter smazsyr writes "An international collaboration of scientists is reporting in landmark detail the decay process of a subatomic particle called a kaon – information that may help answer fundamental questions about how the universe began. The calculation in the study required 54 million processor hours on the IBM BlueGene/P supercomputer at Argonne National Laboratory, the equivalent of 281 days of computing with 8,000 processors. 'This calculation brings us closer to answering fundamental questions about how matter formed in the early universe and why we, and everything else we observe today, are made of matter and not anti-matter,' says a co-author of the paper."
It computes!
Why does this matter?
Blue Gene uses quad core PowerPCs, with 8192 cores on the Argonne system. That's a heck of a lot of days of maxing out your CPUs!
The cesspool just got a check and balance.
Help restore /. to it's former nerd news glory, tag stories like this with realslash to tell the editors that we want our favorite site back.
to the "that's just the way it is" condition. You can go back in time and talk about when the universe created. But then you have to determine the conditions, the rules the mechanics by which that universe was created. Then you have to ask how those rules, conditions and mechanics were created. If you can answer those steps, then you have iterated back just one more layer and will have to answer those questions all over again.
Unless those conditions, rules and mechanics iterate forever, you are forced to a certain point where the universe (or whatever layer you peel back to) just is. Call it "God" or just say that the universe or whatever just is the way that it is.
Can I order a "Raktajino, Hot" from my wall yet?
how is mater formed
how universe get axpadned
They ran the calculation on one core. They needed the other 8191 to render the Aero desktop.
Have gnu, will travel.
The next step in the research will be to determine the remaining unknown quantity that is important to understanding the difference between matter and anti-matter in kaon decay. This last quantity will either confirm the present theory or perhaps, if they are lucky, Blum says, point to a new understanding of physics.
It appears that both theoretically and computationally there is still some work to be done.
When the copyright term is "forever minus a day", live every day like it's the last.
They're not using Pentium III based parallel processing machines.... :-)
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
42. 42 kaons. Ha, ha, ha!
Or is there still insufficient data for a meaningful answer?
ah ha ha ha ha ha ha haaaaaa dick.
(see subject)
-Clio
Karma: Bad (mostly from not giving a fuck)
Blog: http://clintjcl.wordpress.com
Fuckin' Magnets, how do they work?
When all they had to do was ask a 16 year old
FlaccidOS
it's all those damn recursions
Kirk was misquoted. It was "KKAAAOONNN!!!!!"
The article makes sure the readers recognize:
- the breathtaking importance of the subject matter
- the staggering computational resources brought to bear
- the worldwide nature of collaboration between scientists
However, something was missing... like, maybe some actual results that could be summarized in layman's terms?
Brings back memories of an old skit created by frustrated scientists.
Why are they running it on the GPU?
This seemed like a reasonable sig at the time.
All I read about this event is that the computers mapped the decay. Not 1 piece of information about what they learned. In that light, I'll fill in the blanks with the pieces of Quantum Physics I understand.
Kaons are quarks with "strangeness". This typically includes Up, Down, Charm, Strange, and Bottom. Top doesn't participate due to size and shortness of life. Kaons ( http://en.wikipedia.org/wiki/Kaon ) decaying into Pions ( http://en.wikipedia.org/wiki/Pion ) is a great demonstration of quarks participating in the Weak Force. This study combines our study of particle oscillation and weak decay, and digitally maps out that entire process rather than simply relying on theory. Granted, they weren't actually watching this happen, but the generated map gives Physicists what they need to compare against findings from places like the LHC.
TL;DR? Basically, this group designed software and used a very fast computer to generate a result set from theoretical predictions which can be used to compare against various super-collider findings. Specifically, these result sets are regarding Kaon to Pion decay, a Weak force interaction.
Where genius and insanity become confused true wisdom is found
This is a great story, and I'd really like to read the original paper. Does anyone have a link to the original?
It's in the Bible. Now I'm going dinosaur-back riding.
-- I ignore anonymous replies to my comments and postings.
Windows.
Strange title, I don't see an ounce of skeptisisim in your post, but I do see a ton of bitterness and ad-homieniem.
Maybe a bit of pycological projection too? - Are you angry at yourself because you tried to cheat science and it ended badly for you?
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
Since the blog entry contains no reference - and the one hint there is is wrong - here is the actual article reference: Phys. Rev. Lett. 108:141601 (2012) - which was published on 6th April, not 30th March at the article states!
Now onto the physics, sorry but your summary is almost completely wrong. Kaons are mesons which are a bound state of a quark and anti-quark. In the case of neutral kaons this is a strange and anti-down (or vice versa for the anti-kaon IIRC). What is interesting about the kaon is that the neutral states can oscillate between kaon and anti-kaon through a weak interaction. What you end up with is a long-lived kaon (KL) and a short lived one (KS). The simplest way to demonstrate that this system differentiates between matter and anti-matter is to look at the long lived kaon decaying in to muons (heavy cousins of the electron). The number of anti-muons will be about 0.1% different from the number of muons produced.
However the decay to pions is far more closely studied because it can tell us far more information - in particular whether this symmetry breaking occurs in the decay mechanism (direct CP violation) or only in the weak mixing of a kaon to anti-kaon (indirect CP violation). The experiment I worked on as a grad student, NA48, observed this direct CP violation unambiguously for the first time, confirming the previous NA31 result. This ruled out more exotic types of CP violation from a new "superweak" interaction and, in broad terms, was consistent with the Standard Model.
However this was not really confirmation of the Standard Model because the actual calculation of CP violation occurring in the SM is really hard to calculate: it involves quark/W boson loops which must have contributions from all three generations of quarks (specifically including the top quark!). These so-called penguin diagrams (blame the name on John Ellis' dart playing skills!) are really hard to calculate - at least to the accuracy needed for CP violation in kaons. Kaons must decay through a weak interaction because only the weak interaction can change the strange quark into an up quark which is needed for pion decay. However there is also a strong component to the decay.
Strong (QCD) processes are really hard to calculate because perturbation theory does not work for them (the interaction is far too strong). One approach to solve this is lattice QCD which literally simulates all the colour (QCD) fields on a 4D grid of space-time points. However this is really CPU-intensive so only small grids can be simulated. This is not too bad if you have a strong process because, being 'strong' it happens quickly in a small region. However the weak part of the decay occurs more slowly over a larger area. What the authors seem to have done is overcome this simulation problem of both weak and strong forces in the same decay which raises the prospect of accurate calculations of the CP violation in kaon decays which has never been possible before. For the technically minded this paper calculates the Isospin=2 decay amplitude (A_2) whose phase shift, relative to the isospin 0 amplitude (A_0) is what makes direct CP violation visible - it's a really interesting paper - at least if you have ever been involved in kaon physics!
8000 measly processors? Ha! That's a sleepy fishing village. Distributed computing is much more economical and very much more powerful. Of course it is not suited for all kind of work. And people might not join your privately owned, unethically goaled programs, so there is still a niche for traditional supercomputers left...
While we're at it, suggest me a BOINC project worthy of my cycles. Please make a good case, I've been eyeing them all superficially and couldn't decide.
Does this calculation account for the existance of
Dark Matter
Dark Energy
The Dark Side of the Force
The Dark Side of the Moon
I'd mod this 'hilarious', but since I already posted in a few other places I will just comment "LOL" so everyone knows I laughed.
"That's right...I said it."
How is matter formed?
How atom get pregnant?
How is this funny? Why would anyone mod this funny?
I read both the summary and news release and found no useful information. Your comment is the first real news article I've read about this. I don't read physics news stories anymore because they're information free.
stop having fun! .. guys?
If we were made of what we perceive as anti-matter, we'd call that matter and what we think of matter now, would be the anti-matter.
So the reason we are made of matter and not anti-matter is simply a matter of perspective :)
How much is that in bitcoins?
"Honey, what's the matter?"
"You know!"
No, I don't. But maybe maybe a team of scientists using one of the most powerful computers on earth can figure what the heck is the matter with you.
The world is made by those who show up for the job.
One of the big things that originally hooked me was the tendency for people to 'run the numbers' when they had a disagreement with someone else.
Slashdot has always been full of flamewars...
The thing is, years ago it was hardly ever political flamewars. Flamewars about technical matters have an inherent ability for people to point to hard data about things, which kept the whole discussion somewhat tied to reality.
With politics, all bets are off - because you are talking about people with wildly different views about what is good for other groups of people, and even if they agree on THAT you have differences in how to achieve an end-goal. It's all about Seldonesque behaviors of the masses and there's no "numbers" you can run that someone else cannot simply dismiss away with their own numbers.
The reason for the spread of politics here is that inevitably, the spread of technology into the lives of every person means technology gets stuck in the tar baby of political motivation. Technology is simply part of the equation about how to change people in ways you deem most beneficial. So there's no going back to more reasoned discusson unless you want to remove technology from people's lives (some do, but I doubt the motive is to make Slashdot more readable).
It's not like you can make any OTHER site like the "old Slashdot" and have it be any different, due as I said to the intertwining of technology with everyone and politics being everywhere. We all just have to learn how to include politics in technical discourse without getting too heated and off track...
"There is more worth loving than we have strength to love." - Brian Jay Stanley
...was open the good book to find that God created the Universe. Finding this fact need only a couple of seconds of processing time. Silly scientists!
You know that the latest supercomputers use GPUs due to their better paralellisation right?
They are sadists, obviously.
The calculation in the study required 54 million processor hours on the IBM BlueGene/P supercomputer at Argonne National Laboratory, the equivalent of 281 days of computing with 8,000 processors.
But what's that in rods per hogshead? Or is "8,000 processors" some kind of conventional metric for processing?
systemd is Roko's Basilisk.
Comment removed based on user account deletion
why we, and everything else we observe today, are made of matter and not anti-matter
Call me crazy, but I bet that if we and everything we observed were made of anti-matter, we would just call it "matter". :p
Seriously, though, doesn't it have to be one or the other (since a mix will lead to annihilation)? I'm assuming the real question is why what we call "matter" managed to beat out anti-matter instead of a balance of both kinds being made at the beginning, which would then annihilate.
DNRTFA.
"Empathise with stupidity, and you're halfway to thinking like an idiot." - Iain M. Banks
Uh, that idea isn't testable. Seriously, the idea that falsifiability is the demarkation between science and non-science is no more falsifiable than an infinity of possible worlds.
Moreover, if you look at the history of how science is done, many of the singular advancements in science (the Copernican turn, relativity, et cetera) were accomplished in a fashion that paid no attention to falsification. Take Galileo as an example. HIs theories were trivially falsifiable by his own observations. Yet he continued on in a counter-inductive fashion, ignoring the evidence.
And yet the entire article does not contain a single equation, much less a link to the paper. I am disappoint.
May the Maths Be with you!
"Funny" == "herp derp lol!"
If it takes 54 million processor hours to compute it, how long is it going to take for scientists to EXPLAIN it? And who is going to check the results?
Imagine if things had turned out the other way--everything made of antimatter instead of matter.
"...54 million processor hours on the IBM BlueGene/P supercomputer at Argonne National Laboratory, the equivalent of 281 days of computing with 8,000 processors..."
Please, please tell me that the answer was 42.
Now how long will it take for us to compute the question?
-Styopa
Maybe they can compute wtf dark energy is.
We don't have to explain ourselves to this person who believes science should only be pursued for its applications. Basic science paves the road for a wonderful engineering potential, but that's not why we do it.
We did it because SCIENCE IS FUCKING COOL.
Newton and Einstein didn't discover what they discovered out of some search for profit, they weren't Thomas Edison; they thought this science thing was the coolest shit ever and were invigorated by the challenges they offered. Please, on appeal to all scientists, put on your big smile and bend over backwards at fundraisers, but that's not why we do science.
The true intellectual places curiosity and discovery as a virtue unconditionally. It is not to be squelched because it fails to be immediately profitable or applicable.
Why are they running it on the GPU?
Actually a lot of lattice QCD calculations are run on graphics cards, usually NVidia cards like the Fermi and Tesla. These cards have error correction, larger RAM and better support for double precision, but some of us do run on clusters of standard gaming cards like the GTX480.
While the authors (as they always do) consider this landmark, I was unable to find any comment on their letter or the preprint (apparently this) in the usual places. This could be in part because it is a) not 'real' and b) doesn't have the words 'Higgs' or 'superluminal neutrino' in the title.
Interestingly, this calculation (http://arxiv.org/abs/1111.1699) has very little to do with 'How matter is formed'. As the authors clearly state in the abstract,
the calculation is in good agreement with experiment. This currently leads us to believe that its in good agreement with the Standard Model which in turn
means that we have NO idea where the required CP violation for the observed matter/anti-matter distribution comes from.
Their calculation sure is impressive and is in fact the culmination of more than a decade of research into this particular process.
how is babby formed?
We're already made of anti-matter. Except that, since we're here, and we're making the rules and naming things, we call it matter and all the opposite stuff is anti-matter.
I thought that was obvious.
Is it just my observation, or are there way too many stupid people in the world?
Nice result... any plans for independent verification? Cause that's kind of how science works.
This story recalls a thing I read recently:
Someone lamented that many experiments were becoming so hard to duplicate (LHC-results, for example), or require such extraordinary equipment, that there are only 3 or 4 places on the planet that could possibly check it. *IF* they were willing to spend the time / effort / money necessary to verify already published results...
Don't recall the details, but ... yeah. Can we take someone else's word for it?
Because BlueGene is a monumental leap above hol-er-ith cards used during the second world war.
LOL that's why.
I need to know how is babby formed?!
http://undecidedgames.blogspot.com
Next time, just pray your post onto /. and see how that works out for you.
kurzweil_freak
5th Kyu Genbukan Ninpo/KJJR student
Be the darkness that allows the light to shine.