Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Supercomputing, There's an App For That

samzenpus posted about 4 years ago | from the in-the-palm-over-your-hand dept.

Supercomputing 66

aarondubrow writes "Researchers at MIT have created an experimental system for smart phones that allows engineers to leverage the power of supercomputers for instant computation and analysis. The team performed a series of expensive high-fidelity simulations on the Ranger supercomputer to generate a small "reduced model" which was transferred to a Google Android smart phone. They were then able to solve engineering and fluid flow problems on the phone and visualize the results interactively. The project proved the potential for reduced order methods to perform real-time and reliable simulations for complicated problems on handheld devices."

cancel ×

66 comments

Sorry! There are no comments related to the filter you selected.

appx. (0, Troll)

swanzilla (1458281) | about 4 years ago | (#33294476)

Approximating supercomputing is not itself supercomping, fellas.

Congrats on making /., however.

Re:appx. (4, Insightful)

flaming error (1041742) | about 4 years ago | (#33294522)

It sounds like the supercomputer generated an algorithm for the smartphone to run. I guess they can call that "leveraging the power of a supercomputer" but implying the phone app is doing supercomputing stretches things a bit far. I call misleading headline.

Not even... (3, Insightful)

Monkeedude1212 (1560403) | about 4 years ago | (#33294568)

The team performed a series of expensive high-fidelity simulations on the Ranger supercomputer to generate a small "reduced model" which was transferred to a Google Android smart phone

This is like saying that watching Toy Story on your iPhone leveraged the massive renderfarm used by Pixar.

Re:Not even... (3, Informative)

Ambitwistor (1041236) | about 4 years ago | (#33294636)

Not really. In this case, the smart phone isn't simply rendering output of a supercomputer simulation.

Re:Not even... (1, Insightful)

Anonymous Coward | about 4 years ago | (#33294708)

This is a university PR piece, and these are notorious for being vague about what they are actually claiming .

It sounds like the supercomputer is basically doing all the grunt work and the phone is doing something analogous to interpolating the results. For example if the supercomputer supplies pre-computed results for some question for parameter alpha=1.0 and alpha=2.0 and if the user selects alpha=1.5 then the phone will interpolate the two supplied results and get an answer that will (if the interpolation method is a suitable choice) be a very accurate approximation to the full calculation for alpha=1.5.

It's more than just rending, but perhaps not that much more.

Re:Not even... (2, Interesting)

Ambitwistor (1041236) | about 4 years ago | (#33295466)

What the phone is doing is "reduced order modeling", which (if the article is using the term accurately) finding a simple set of equations whose solution provably approximates a far more complex system of equations. It's not just interpolation (lookup tables, or machine learning from a training ensemble). Reduced order models actually have dynamics in them.

The potential innovation here, as I see it, is that it takes some supercomputing effort to build the reduced order model for a specific problem. So you can input a description of your problem into the phone, communicate it to the supercomputer and let it analyze the problem for a couple hours, and it spits out a custom approximation to your problem that you can run in the field on your phone.

Now, what I'm not sure that they've developed is some convenient way to specify the problem using the phone's interface. Maybe the problems they're working with are too complex to be easily specified with a few keystrokes. If all they're doing is loading a precomputed reduced model onto the phone, and there's no interaction with the supercomputer to let it handcraft solutions to problems in semi-realtime, then it's not nearly as interesting.

Re:Not even... (3, Informative)

Ambitwistor (1041236) | about 4 years ago | (#33295510)

I just re-read the article. What it sounds like they're doing is having the supercomputer craft a reduced order model which is optimized to a particular range of parameters. That suggests to me that they're constructing a perturbative model about some fixed solution that the supercomputer produces. Perturbative approximations are more accurate the closer they are to the "reference" solution. So the innovation appears to be: the user can specify what set of parameters they want to perturb about, and therefore construct a custom model which is optimized to perform well in the parameter range that user is interested in.

Re:Not even... (1)

somersault (912633) | about 4 years ago | (#33294748)

More like doing a rasterised live render of Toy Story rather than watching the fully ray traced version.

Re:Not even... (1)

mdwh2 (535323) | about 4 years ago | (#33300424)

Don't worry, that's queued for tomorrow's story: "Watching Toy Story On The iPhone(TM)".

(Certainly as newsworthy as many stories we get, such as yesterday's "Choose your own adventure On The iPhone(TM)" or some time ago was "You can access this website On The iPhone(TM)".)

useless (-1, Flamebait)

Locutus (9039) | about 4 years ago | (#33294490)

they did it on the phone nobody bought, the Nexus One.

It's funny how a phone which didn't sell well seems to keep showing up in press releases. At least it looks good enough to use in press releases and super computers so it's got that going for it.

LoB

Re:useless (-1, Offtopic)

Locutus (9039) | about 4 years ago | (#33294500)

oh, and first post!

LoB

Re:useless (-1, Offtopic)

Anonymous Coward | about 4 years ago | (#33294558)

Missed it by THAT [holds finger and thumb together half an inch apart] much!

Re:useless (1)

AaronLS (1804210) | about 4 years ago | (#33294670)

I would make the educated guess that it could easily work on other Android platforms, not just the Nexus One, depending on their implementation.

Re:useless (1)

somersault (912633) | about 4 years ago | (#33294770)

oh, and first post!

Well, approximately.

Butt does it hurt? (1)

symbolset (646467) | about 4 years ago | (#33296942)

Does it still hurt that the G1 moved more units in its first minute in the market than the Microsoft Kin did in whole lifespan? That would be nice if it bothered you. I subsist from your pain. I'm a pain vampire that way.

The Google phone is still available through other vendors though not from Google, but it did what it was intended to do: convince vendors that there was a market for the Android candybar phone. It's proved and there's no more need for the proof because the idea has taken off. Google doesn't have to sell that idea any more, and the G1 is still well supported. Now that Android is proven on the candybar, new models with more advanced processors are available that exploit the available app market for that form factor - a market that would not exist if Google had not broken the ice. Google told the market: if you won't make it, we will! And it was popular so the market made it and now they don't need to beat them to death about how it was a good idea. They hoped also to prove the direct market for phones to defeat the subsidized model and that didn't work out so they failed there - but WE still won.

Google isn't and doesn't want to be a phone vendor. Their interest is ensuring that if people search they use Google search. Their interest is in maximizing your bandwidth so you can search more times per day. They would prefer it if your mobile phone defaulted to Google search, because mobile clients yield more cents per ad exposure than desktop clients. But as a fallback position if your browser can render their simple home page and you can set it as your home page they're ok with that because you chose it. That's their whole business model in a nutshell.

Google isn't and doesn't want to be a wired internet provider. But if you had gigabit broadband you would see their ads more, so faster internet benefits them in a way that doesn't hurt them. They're working their gigabit broadband initiative. Same with wireless spectrum, or any of the many other things they do. Sometimes they buy us freedom (with broadband spectrum for example) without committing a cost to themselves or spending any money. Some of this is quite awesome - the sale of spectrum used to be a quiet thing but now that Google's engaged we're aware of how our government is doling out our communal spectrum property, and how we suffer when it's sold to the wrong people.

I'm not saying that Google is angelic here, except in comparison to the incumbent providers of bandwidth. It goes to motivation: Comcast (or whatever provider) is motivated to deliver the least bandwidth for the dollars paid because to Comcast bandwidth costs money and the money you pay is where Comcast's money comes from. Google is motivated to deliver the most bandwidth per dollar paid because to Google more bandwidth equals more ads per minute, because people who hope to sell you stuff is where Google's money comes from. To truly understand this difference you have to understand that you personally are the object that is bought and sold here. Your interests, your motivations, your desires, and most especially your disposable income are the object of this engineering. Properly speaking you should be in favor of the Internet building a path you your wallet unless it's done in a way that's dishonest or unfair. It allows you to buy any object for sale anywhere, when once you had to go to South Africa to buy tribal masks as a travelling trophy. For me personally there are some aspects that make me uncomfortable enough that I look carefully at the terms of the deal.

As a subscriber you can choose to subscribe with a vendor whose interests are apposite of yours, or one whose interests are aligned. It's your choice, and Google doesn't want to limit your choice here but you can be sure Comcast does. Comcast has an established market to protect, much like MaBell did back when we still had landlines (remember those?)

Ultimately people will choose and I respect that. The purpose of my post isn't to run down those who've chosen already, but to hopefully provide more information to people about to choose so that they can make a more informed choice. I have some insight into extraction processes, but each one is custom so if you've chosen already and it's time to escape - reply below or send a gmail to my username.

Re:useless (1)

mdwh2 (535323) | about 4 years ago | (#33300494)

It's funny how a phone which didn't sell well seems to keep showing up in press releases

The Iphone?

(Apple have about 3% of the market, yet get a mention several times a day in any random Slashdot story; to put things in perspective, Nokia ship twice as many phones per quarter than Apple have ever sold, and even just one of their many products, the 5800, has sold equal to or more than the original "iPhone". Android has now already overtaken Apple btw, and is the fastest growing platform, whilst Apple are actually now increasing sales slower than Symbian, Android and RIM, according to Q2 2010 results.)

Re:useless (1)

Locutus (9039) | about 4 years ago | (#33306044)

good point and valid too. But to be fair, the press doesn't write many stories about nuts, bolts and screws( sold in the billions of units ) but they'll write about nice shiny new cars made up of nuts, bolts, and screws. The point is, utility devices are boring to the press but put something shiny in front of them and they'll gobble it up. Not to mention that most in the media segment are of the artsy-fartsy type and therefore are more likely to side with Apple. Remember, Apple users were constantly fighting just to keep their platform of choice as big businesses constantly tried to consolidate on only Microsoft products. I see so much press coverage as some what of a 'fighting back' consequence of so many years of oppression.

As for Android, it's cool and cheap and every where so it's finding press coverage also.

And BTW, my original post was a jab at the press for putting down the Nexus One. I should have been more obvious but I thought more on /. would have got that.

LoB

PR Bullshit (5, Informative)

pigwiggle (882643) | about 4 years ago | (#33294538)

The money quote "This is not the first time that model reduction algorithms have been used to ameliorate the complexities of large-scale physical simulations. The advantage of the system designed by Knezevic and his colleagues is its rigorous error bounds, which tell a user the range of possible solutions, and provide a metric of whether an answer is accurate or not. The error bounds are based on mathematical theory developed in Prof. Patera's research group at MIT over a number of years. "

The research is about error bounds on coarse grained models. The smart phone is just hype.

Re:PR Bullshit (3, Funny)

oldhack (1037484) | about 4 years ago | (#33294624)

Why are they spamming slashdot anyways? Has the academia gotten so pathetic that they deem any exposure is a good exposure?

Re:PR Bullshit (2, Insightful)

blair1q (305137) | about 4 years ago | (#33294848)

Everyone is spamming slashdot, and the people voting on the firehose are generally too lame to understand it. Throw "reduced order methods to perform real-time and reliable simulations" at them and they click the + just to look smart.

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33306158)

Why are they spamming slashdot anyways? Has the academia gotten so pathetic that they deem any exposure is a good exposure?

Not academia, but the people who pay for it. They've started demanding lots and lots of "outreach". It even plays a role into the decision on funding a proposal. Researchers with good outreach get better staff reviews and hence promotions and pay raises. PR has always paid off, but it's become part of the system to encourage it now.

Re:PR Bullshit (3, Funny)

WankersRevenge (452399) | about 4 years ago | (#33294626)

I didn't realize you could run algorithms on a smart phone ... these guys are brilliant! ;)

Re:PR Bullshit (5, Funny)

blair1q (305137) | about 4 years ago | (#33294882)

Go into settings and select "scientific mode". That makes a lot more buttons appear.

Re:PR Bullshit (1)

interval1066 (668936) | about 4 years ago | (#33295264)

And as we all know, just like on a general's uniform, more buttons means more power.

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33298686)

But no square roots

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33299512)

Toggle Inv and use x^2

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33294780)

I agree with parent. The research may be interesting to people within the area of "model reduction." However, this is a bit esoteric for the front page of slashdot. But, people get a boner if it says MIT.

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33295550)

OH gosh. Please warn us in advance if you're going to throw around acronyms like that in a comment. That was almost really embarrassing.

Re:PR Bullshit (1)

CarpetShark (865376) | about 4 years ago | (#33294838)

The research is about error bounds on coarse grained models. The smart phone is just hype.

And if there research gets lost, the credit claimed by someone else coming along later and doing the same thing again, all because they obscured their research behind misleading buzzwords? Tough.

Exactly (1)

Anubis350 (772791) | about 4 years ago | (#33294846)

I had images in my brain of an article about getting torque and mpi (something that, to my knowledge, there's no reason outside practicality to stop from working) plus some sort of auto-meshing running on android phones or some such (though using wifi as an interconnect makes me shudder), *that* would be phone supercomputing, this is *not*.

Re:PR Bullshit (1)

ygtai (1330807) | about 4 years ago | (#33295268)

The smart phone is not totally a hype. It serves kind of like a proof of concept or a demo. If the reduced model can be run on a smart phone quickly and accurately enough, it can be run on similar embedded devices. This could possibly be commercialized pretty quickly.

Re:PR Bullshit (0)

Anonymous Coward | about 4 years ago | (#33297794)

"Concluded Knezevic: 'When you tell people you can solve a problem that would normally take two hours on Ranger in one second, with guaranteed error bounds, people instantly understand what model reduction is all about.'"

From the article.

Yes, it's PR. This research (reduced model methods) are still freaking cool. To me. Of course that's probably because I can think of lots of applications for this (substitute embedded device for smart phone). But I'm not an IT guy, I'm an engineer.

Re:PR Bullshit (1)

sjames (1099) | about 4 years ago | (#33307094)

Yep. The research is good, but the smartphone is just an angle. Sorta like if I write an adventure game ans say "it could be the first text adventure played on a space station" just because it isn't actually incompatible with the laptops on ISS.

What? No Windows Mobile? (0)

Anonymous Coward | about 4 years ago | (#33294574)

I'm surprised they didn't use Windows Mobile for this :-)

Wow. (1)

pushing-robot (1037830) | about 4 years ago | (#33294602)

So... if you analyze a problem and discover you can get mostly accurate results from a simple algorithm, you don't need a supercomputer anymore? What a concept! I'm going to go write the first physics simulator for personal computers!

Seriously, the cool bit is that they're generating these reduced models programatically. But the way it reads, it sounds like the reduced model itself, and the fact that it runs on smart phones are the important parts.

Re:Wow. (1)

khallow (566160) | about 4 years ago | (#33294652)

Seriously, the cool bit is that they're generating these reduced models programatically. But the way it reads, it sounds like the reduced model itself, and the fact that it runs on smart phones are the important parts.

And that sounds accurate to me. It's a demonstration, among other things, that you can control a complex system, that originally required the efforts of a supercomputer to model, with far simpler tools.

So (1)

jewishbaconzombies (1861376) | about 4 years ago | (#33294622)

Is that a supercomputer in your pocket or are you just glad to see me?

Re:So (1)

blair1q (305137) | about 4 years ago | (#33294898)

I remember when 100 MFLOPS was munitions-grade computing. Now you have 1 GHz in your pocket and 4 cores of 3 GHz hooked up to 450 cores at 1.5 GHz and you still think Crysis is stuttering even in low res...

Semantics, maybe... (2, Informative)

mea37 (1201159) | about 4 years ago | (#33294638)

...but I'm going to go ahead and argue that they are not "performing supercomputing on a phone", because that kind of marketing doesn't belong in research.

Yes, it could be very useful; I have no doubt it's just as useful as they claim. And yes, it allows someone in practice to solve a problem "in the field" with a phone, when otherwise a supercomputer might have to be used.

But the supercomputing was done on a supercomputer in advance, when the reduced model was calculated. Its just that instead of giving one specific answer for one specific input, the supercomputer is returning an algorithm that will approximate the answer within known error bounds for a specified domain of inputs. Executing the algorithm isn't supercomputing (if it were, you couldn't do it in a few seconds on a phone); it's using the fruits of the earlier supercomputing that produced the algorithm.

Re:Semantics, maybe... (1)

hesaigo999ca (786966) | about 4 years ago | (#33303888)

I tend to agree, as I could also use my smartphone terminal to log unto that supercomputer, and have a thin client application that is the GUI for setting most info for the processing needed to be done, then let the supercomputer do its thing, and then have that thin client visually display the results, so no real supercomputing going on here...

Well, now ... (0)

Anonymous Coward | about 4 years ago | (#33294642)

I'll sleep better at night knowing my pr0n is finally safe from my wife..

*sings* (1)

Picardo85 (1408929) | about 4 years ago | (#33294658)

he's got the whole world in his hand ...

marketing (1)

spirito (1552779) | about 4 years ago | (#33294738)

This is just marketing at work. from TFA: The real impact of the system may come in the application of these methods to aircraft or automobiles, which use control systems to react to inputs from the environment in order to achieve optimal safety and performance. Examples include traction control in cars and stabilization systems in jet fighters. “If you have sensors feeding in data to the reduced order model system, then it could solve the equation corresponding to the input data, and indicate the appropriate response in real-time based on the calculations you performed on a supercomputer,” This is how things work already: control systems on a jet fighters do not solve a CFD problem to know how to control the plane, they have a built in model (yeah, "reduced order", if you want to call it this way) that approximate the actual behavior of the plane. Doing it on a smartphone is useless. Furthermore the article has no details on how the error bounds are claculated.

Re:marketing (1)

blair1q (305137) | about 4 years ago | (#33294940)

It's a table lookup where the things you look up in the table are easily-solved algebraic equations across the small domain of the single table entry, instead of a continuous model in differential equations across the entire state space of the table.

The thing that can solve the continuous model in differential equations across the entire state space is a supercomputer. The thing that can chop the continuous model up into a table of simple algebraic equations is also a supercomputer. The thing that can look simple equations up in the table and solve them needs only about the power of a smartphone.

And it's not unlikely it can do it in real-time, but I'm pretty sure nobody's done the testing to do it in real-time in a safety-critical system, so implying that you can use your smartphone to control anything in your car other than the entertainment system is completely naff.

Re:marketing (1)

Ambitwistor (1041236) | about 4 years ago | (#33295544)

Yes, we know that reduced order models already exist. The new thing here is that you can input the parameter range you want, wait a few hours, and the supercomputer sends you back a custom reduced order model optimized for the parameter range you care about. You can then apply this model "in the field" to the situation you're dealing with. It's supposed to be useful for situations where the details of the problem aren't known ahead of time, and you can't pre-compute the reduced order model.

And no, the article doesn't have details on how to calculate error bounds, because it's just a news article. The guy has been publishing on this algorithm for years. I'm sure you can look up how the errors are calculated if you want.

Re:marketing (2, Informative)

jwpeterson (1299277) | about 4 years ago | (#33304944)

Furthermore the article has no details on how the error bounds are calculated.

Good point, if you are interested in the details of the error bounds, please check out our preprints below, and the references cited therein.

http://www.tacc.utexas.edu/~peterson/articles/2010_rboomit_cmame_preprint.pdf [utexas.edu]
http://www.tacc.utexas.edu/~peterson/articles/2010_hafs.pdf [utexas.edu]

Put CUDA on a phone, then we can talk (2, Insightful)

jpapon (1877296) | about 4 years ago | (#33294774)

I'll believe they've created mobile supercomputing when someone puts a powerful GPU that is CUDA-ready in a smartphone.

Of course, you better get some big batteries for your phone, because Teraflops ain't free

Re:Put CUDA on a phone, then we can talk (2, Funny)

blair1q (305137) | about 4 years ago | (#33294952)

CUDA isn't supercomputing. CUDA is more like super-doopercomputing. And it's a fucking crime that nVidia isn't doing better in the market with it.

Re:Put CUDA on a phone, then we can talk (2, Interesting)

Entropius (188861) | about 4 years ago | (#33295696)

The lattice QCD people, at least, are porting their code to CUDA just as fast as they can. The bottleneck right now is that there's no good way to get multiple GPU's to communicate (quickly). So, for the largest problems (simulating a 64x64x64x192 lattice), you still need a conventional supercomputer (like Ranger, the one in the article here), because it's just too huge to put on a single GPU and multi-GPU doesn't scale well.

But for smaller problems (like a 24x24x24x64 lattice), GPU's will be great, and people are developing this capability as fast as they can.

Yes smartphones can display results from (1)

BitZtream (692029) | about 4 years ago | (#33294908)

super computers, big whoop.

The phone isn't actually doing anything but functioning as a UI.

Re:Yes smartphones can display results from (2, Informative)

Kurofuneparry (1360993) | about 4 years ago | (#33295148)

That's an RTFA comment right there.

This isn't just a UI, it's a reduction of the algorithm provided by a supercomputer. However, I believe that this first set of lines is misleading, inaccurate, and likely an example of the writer not knowing what they're talking about:

What if you could perform supercomputing calculations in real-time, on your smartphone ... Researchers ... have created an application that does just that.

It doesn't do supercomputing because it isn't a supercomputer, it just makes an educated guess based on sitting at the supercomputer's knee and playing "monkey see, monkey do". Not a bad trick but the claim's overwrought.

Re:Yes smartphones can display results from (0)

BitZtream (692029) | about 4 years ago | (#33295604)

I'm pretty sure you have a different definition of 'supercomputing' than the rest of the world and I.

I'm fairly sure you also don't understand what they are talking about either. They are attempting to imply they are doing the work of a super computer on the phone, and in fact the super computer is doing the brunt work of reducing it so it only leaves a tiny bit of work left for the phone to process so it looks impressive. That tiny bit just happens to be the most useful bits for a person to play with.

I don't care what you call the algorithm, I don't care what name you come up with to describe what they are doing as if its something different than the same shit we've been coding for years.

You can perform super computer calculations in real-time on a TI-85. Just like on your Android phone, the calculations run in real-time on both. The only way to not run them in real time is to use an emulator. They didn't say there were simulating in real time.

You have been scammed. You bought into a precisely worded press release designed to trick you into thinking it was something new, exciting and revolutionary.

10x10 = 100 on a supercomputer, a calculator and an Android phone. Math is math regardless of where its done. There is no calculation you can run on a 'super computer' that can't be run on an Atari 2600 assuming you gave it access to enough RAM.

The defining factor for 'supercomputing' is that is does A VERY VERY LARGE AMOUNT of calculations in tiny amounts of time. More realistically it means using more processing power than is generally easy to gain access too, so its only the fastest of the fastest that get considered to be super computers.

Most of the work is done on a super computer, then a tiny UI layer is thrown on top to play with one small tiny part of it and tweak the calculation results and display them.

And please ... don't ever use the phrase 'educated guess' as if its a good thing for a computer to do. An 'educated guess' means you are in fact ignorant of information required to make an accurate calculation. Thats not really all that useful unless you're doing stock market speculations or selling home mortgages

Re:Yes smartphones can display results from (1)

Entropius (188861) | about 4 years ago | (#33295710)

If you replace "educated guess" with "first-order approximation", it sounds a lot better -- and, in fact, this happens in the sciences all the time. But that's just what a first-order approximation is: it's a guess (based on the first term of a series) that is educated (based on some belief that the subsequent terms are smaller).

Re:Yes smartphones can display results from (1)

Kurofuneparry (1360993) | more than 3 years ago | (#33340498)

Wow, must have hit a nerve. Where to start:

I assure you, I understand supercomputing well enough and that your comments on supercomputing don't counter what I was saying. Your description of the action of their software:

Most of the work is done on a super computer, then a tiny UI layer is thrown on top

doesn't accurately describe model reduction algorithms. Another commenter pointed out that 'first order approximation' is a better term. The phone doesn't 'finish the last bit' of processing but makes a low-order approximation of the entire process.

As Entropius also noted, educated guesses (first order approximations) are VERY common in science, medical research and many other areas. They quite often work great. I do say 'educated guess' as if it's a good (enough) thing for a computer to do. In TFA, they state that the results are returned with error bars for the estimation, making the results quite potentially useful.

Again, I'm not 'scammed' because I realize that their claim is overwrought, but still this is a potentially useful method. They're not the first to apply the idea but it's still interesting and more than just a UI.

FEMM for android. (2, Insightful)

Facegarden (967477) | about 4 years ago | (#33294926)

I've been using FEMM lately for some magnetics stuff I've been working on. I would LOVE an android port, or some way to run simulations from my phone.

I don't *really* need it, but its just funny how something like that is actually possible these days. We probably will have supercomputers in our hands someday. I mean, current phones already are supercomputers by the standards of what...? 30 years ago? 20 years ago?

Smartphones will become the tricorders of the future, its inevitable.
-Taylor

Re:FEMM for android. (0)

Anonymous Coward | about 4 years ago | (#33295172)

Smartphones will become the tricorders of the future, its inevitable.
-Taylor

There's already an app for that, and it isn't just a useless toy (I thought it was when I downloaded it and was pleasantly surprised.)

Re:FEMM for android. (1)

Failed Physicist (1411173) | about 4 years ago | (#33297958)

My stock Nexus One running Froyo (Android 2.2) gets an effective 34 MFlops on linpack (and I didn't even kill the background tasks).
This is better than 1969's top supercomputer (theoritical peak of 36 MFlops, effective ~10) and equivalent to 1974's top supercomputer CDC STAR-100 [wikipedia.org] that had a theoritical peak of 100 MFlops but which had much lower realworld performance.
The nexus one cost me 600 with tax and shipping. The CDC 7600, which is easily beaten, cost 5 millions in 1970's dollars.

Run-of-the-mill modern desktops are much more impressive; a 300$ nVidia GPU can easily push a 800$ desktop (total) over 1 TFlop, something which supercomputers only achieved back in 1997 with Intel's ASCI Red 9632. [wikipedia.org]

Oh and btw Android already runs a very nice app called Tricorder. It's already here ;)

Re:FEMM for android. (0)

Anonymous Coward | about 4 years ago | (#33298864)

I would LOVE an android port, or some way to run simulations from my phone. I don't *really* need it, but its just funny how something like that is actually possible these days.

SSH has been around since '95. And Telnet since '69.
And I routinely do log into supercomputers to manage/check simulations from whatever ssh enabled gadget/computer is within reach, including "smartphones".

Nice one MIT: I have Prior art (0)

Anonymous Coward | about 4 years ago | (#33295160)

I had a supercomputer performing simulations (ground water) and was able to visualise the results interactively on a mobile phone back in 2004.

Partial Writeup [lockyerwater.com]

Super computer access (0)

Anonymous Coward | about 4 years ago | (#33295690)

Boy, you can use Python and solve for primes for days tapping a super with this device.

Batteries, anyone? (1)

yagerd001 (857569) | about 4 years ago | (#33295910)

Great. Another way to kill my Nexus One's battery. There's a reason I always keep two fresh ones in my pocket (batteries, you perv).

Seems inefficient (1)

SpaFF (18764) | about 4 years ago | (#33296290)

Supercomputers are big. Even when idle they still require lots of power and cooling, so ideally you want your supercomputer to be 100% utilized all of the time. That's why most supercomputers are "over-subscribed" and have batch schedulers (moab/torque, PBS, LSF, etc.). Users submit jobs, and the scheduler goes about placing those jobs on the supercomputer in a way that keeps utilization as close to 100% as possible. This means that typically when you submit a job it will not run immediately.

If your cellphone "out in the field" is relying on a supercomputer to do calculations, you probably aren't going to want to sit there waiting the minutes/hours/days it might take for your job to make its way through the queue. So you have a few choices: Make some sort of system reservation and only use your phone during the reservation time (probably not practical when you are "out in the field"), configure your scheduler to pre-empt currently running jobs in favor of the "cell phone" jobs (this might piss off non-cellphone users), or dedicate some or all of the system to doing nothing but being available for cell phone jobs....and the portion you dedicate will have to be enough to cover all of your cell phone users.

The last option is probably the best in terms of making sure that there is always supercomputing resources available for the cell phone users, but this undersubscription will cause your supercomputer to sit idle when field work isn't being done. So suddenly you are paying to power and cool a supercomputer that is sitting there waiting on the user to do something.

Supercomputer companies are slowly working on making supercomputers "greener", i.e. requiring less power/cooling, the ability to power off cpus/nodes/frames when not in use, etc. But until this green technology is perfected paring supercomputers with cell phones seems like a very inefficient way to do things.

Re:Seems inefficient (2, Insightful)

Bill Barth (49178) | about 4 years ago | (#33299856)

You don't understand how this works. You do the computation ahead of time on the supercomputer to build your reduced order model which you download onto your phone and take out into the field. Once you've downloaded the model, you don't need the supercomputer any more. You can use the phone to do computations using the reduced model as much as you like. If you get into a regime where the predicted error from the reduced order model is too high, you can go back to the supercomputer and update the model. If that happens, then you'll probably have to wait in queue again, but that's not such a big deal.

as soon as Cray has a smartphone, I'll try it (1)

swschrad (312009) | about 4 years ago | (#33296602)

oh, wait, it will probably be on (shudder) ATT. all 11,000 pounds of freon-cooled 3-phase phone.

Mobile Objects (1)

Doc Ruby (173196) | about 4 years ago | (#33297156)

I've been waiting over a decade for Java's features to support mobile objects to have an infrastructure that made deploying them worthwhile. Why send the logic around the network, instead of just sending the data to where the processors are? Well, with the vast majority of computing power now distributed among so many users, and mostly idling across the year, it's worth using distributed supercomputing now. Folding@Home was a good start, but the distributed app should be generic enough that any crunching can exploit whatever processing it can get access to.

Now we need a marketplace that can aggregate all that processing to consumers of it, and compensate the people who let the mobile objects run on their devices. The natural operators of that system would be the telcos. But since they never do anything new, they'll be too late to the game to be the ones who do it. So who will?

If I were Larry Ellison, I'd have Oracle do it. And sell more Oracle machines to run the Java infrastructure.

web access (1)

virtuosonic (1880050) | about 4 years ago | (#33306250)

why didn't they simply created installed an apache server and access the result trough http using the web browser of any phone not just an android
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>