Forgot your password?
typodupeerror
Software Biotech PlayStation (Games)

Folding@home GPU2 Beta Released, Examined 149

Posted by Zonk
from the nice-to-give-my-console-a-hobby dept.
ThinSkin writes "Stanford has recently released an update to their Folding@home GPU-accelerated client, which includes notable upgrades such as support for more current Radeon graphics cards and even a visualizer to see what's going on. ExtremeTech takes a good look at the new Folding@home GPU2 client and interviews Director Dr. Vijay Pande about the project. To the uninitiated, Folding@home is a distributed computing project in which hundreds of thousands of PCs and PS3s devote a portion of their computing power to crunch chunks of biological data. The goal is 'to understand protein folding, misfolding, and related diseases.'"
This discussion has been archived. No new comments can be posted.

Folding@home GPU2 Beta Released, Examined

Comments Filter:
  • by dreamchaser (49529) on Wednesday April 16, 2008 @04:39PM (#23095408) Homepage Journal
    Just think of all the global warming caused by all those CPU's and GPU's cranking away day and night! And all that electricity used! The horror! They are making Al Gore cry!

    (I am joking, for those of you who are humor impaired)
    • by JeanBaptiste (537955) on Wednesday April 16, 2008 @04:44PM (#23095456)
      you should start a StopGlobalWarming@Home project, where spare CPU cycles go towards global warming research.
    • Re: (Score:2, Funny)

      by lobiusmoop (305328)
      That's the price we way for trying to see God's face. We're building a modern-day Tower of Babel with our tech in our quest for the singularity.
    • Re:Global Warming! (Score:5, Informative)

      by vux984 (928602) on Wednesday April 16, 2008 @05:10PM (#23095738)
      You shouldn't be joking.

      Folding @ Home on a PS3 costs the average participant around $150-200 year in electricity if they run it 24x7. Up to $400+ in places where electricity is more expensive. PCs average less, but only because so many of them are lower power, while all PS3s are high wattage.

      I think its a worthwhile project, but the electricity people are donating isn't free and F@H uses a lot more electricity than most people think. "Oh, I've got my PC on anyway", or "Oh it can't be as much as my fridge." both of which are mistaken, your fridge uses a fraction of what a PS3 running F@H does, and even if your PC is on, running at idle or going to sleep uses a LOT less power than maxxing out the cpu and/or gpu 24x7.

      A PS3 running @ 280W 24x7 for a year:

      280W x 24h/d x 365d/y = 2452800 Watt-hours/year or 2452 kWh/y

      at $@.12/kWh that'll cost you: $294.00 / year

      Then multiply that by the number of PC's running it... it adds up fast.

      Like I said, its a good program and a good cause, BUT its not free. A kid/teen shouldn't be running it without a parents permission and understanding of the cost.

      I don't like the F@H 'propaganda' because I think its somewhat deceptive about the costs. Its relying on peoples attitude that their free cpu time is truly free to prevent them thinking about the real costs. If you probe they don't lie about the costs, but ethically they really should be more upfront about them.

      And now that there is money involved, I should choose the best use of it. When I'm faced with a decision of choosing the best place to donate $300 I think their are other causes more worthy of my money than F@H. But that's a personal choice. If you want to donate to F@H, by all means do so.

      One final issue - generally when you donate more than $10-20 to charity you get a tax receipt. $150-500 quite a bit more than $10.
      • Re: (Score:2, Insightful)

        by boombaard (1001577)
        hm.. this way you're directly investing in 'new' science, and you know what the goal is.. if you invest in amnesty/OxFam/whatever you know at least 20% is lost due to "overhead", another 10% at least is lost due to corruption, and even then (in the case of oxfam and related charities), there is a chance you're funding an organization that has more than a few members (statistically speaking, based on the amount of cases that have come out over the past 5 years or so) that indulge in sex-for-food programmes w
        • Re:Global Warming! (Score:4, Insightful)

          by vux984 (928602) on Wednesday April 16, 2008 @08:27PM (#23098172)
          hm.. this way you're directly investing in 'new' science, and you know what the goal is..

          Fair enough. But its a little dishonest if you don't REALIZE how much you are invested. That's my biggest issue. Once people know what it costs I have no issue if they're still willing to contribute. But it bugs me, especially since I beleive the a very significant proportion of the people contributing to F@H are not the one's paying the bills.

          The other part is how much do F@H results actually cost, in aggregate? Is it good value for the science produced? They've consumed between $50 and 100 million in electricity. Could they have made better progress towards their goals if they were given the money directly? At the very least if they built their own super computer and managed the costs directly the waste would be far far less.

          Not only would they be paying industrial rates for electricity instead of residential rates, they'd also be using far less of it because they'd have racks of CPUs not all powering hard drives, and what not needlessly.

          Hell, just take a look at the from their site: (For the purposes of this I've assumed that it costs 'volunteers' on average $10 to run a cpu per month in electricty.)

          190,000 PCs generating 182 TFLOPs. 191k cpus. Total Cost ~1.9M/month. ~$10,494/TFLOP/month
            41,000 PS3 generating 1257 TFLOPs. 41k cpus. Total Cost ~0.4M/month. ~$326/TFLOP/month

          What moron would keep the PCs running?

          A final note about overhead. You lose 10-20% efficiency right off the top with F@H due to the lack a tax receipt. I can donate $250 to a registered charity at the same cost to me as buying $200 worth of electricty due to the taxes. Or conversely when you donate $200 to F@H -you- pay an extra 20-50 in taxes vs had you given the same $200 to a registered charity.

          but if i had to choose, and if i had a choice, i'd rather invest in an @home project.. i find it a lot more intrinsically motivating than knowing i'm keeping a statistic alive that in 10-20 years might start earning their country some money through taxation because he's had his K-6 education.

          Between those two I'm inclined to agree. I tend to mostly donate to small local organizations myself.

        • by afidel (530433)
          I don't know, your numbers are pretty pessimistic. The audited numbers for my local food bank are 6% administrative overhead and 3% fundraising. From the numbers I've seen online that's pretty typical for US foodbanks.
      • Re: (Score:3, Interesting)

        by dreamchaser (49529)
        My tongue was only half in my cheek. I stopped running any and all distributed clients a couple of years ago precisely because of the resultant power/CPU utilization. I didn't do it for the environment though as I alluded to in my joke. I did it to save money on my electrical bill.
      • Re:Global Warming! (Score:4, Informative)

        by SecondHand (883047) on Wednesday April 16, 2008 @05:55PM (#23096180)
        It seems that the PS3 40 GB consumes only half of what you said (135 Watts, see http://www.engadget.com/2007/10/30/40gb-ps3-features-65nm-chips-lower-power-consumption/ [engadget.com]).

        So you can go and buy a second PS3.
        • Re:Global Warming! (Score:4, Informative)

          by vux984 (928602) on Wednesday April 16, 2008 @06:34PM (#23096734)
          Thanks! You are right.

          There was information when the PS3/F@H launched that consumption was 280-300W, but apparently that was actually around 200-220W so my post above was out by ~$70, and now with the newer lower wattage PS3s the price comes down even more.

          But even at 135W, assuming the same .12c kWh I used in my original post that's ~$150/year. Maybe not a big deal to some, but how many would still sign up if they had to pay $150 to f@h directly instead of having it nickle and dime them daily on their power bill? I suspect the user count would be orders of magnitude lower.
      • Re: (Score:2, Interesting)

        by dstates (629350)
        Of course there are also all of those Flash ads that continue to run even when the browser tab that they are on is not visible. They continue to consume CPU and electricity so they are also adding to your power bill. You think I am joking, but if you are like me, you may have a dozen tabs open at any given time and each of those pages may have several active graphics items on them. Adds up.

        Good reason to run FireFox and AdBlock or FlashBlock. Even better, turn your PC off when you are not using it.
      • by evanbd (210358)

        One final issue - generally when you donate more than $10-20 to charity you get a tax receipt. $150-500 quite a bit more than $10.

        The last time I donated to charity (clothing, not money), I got a receipt that said I had donated, but not how much -- I was responsible for filling in the details and providing any documentation of value I needed. If that's acceptable for Good Will, etc,., it should work for Folding@Home. They don't need to come up with a dollar figure, you can do that. They already tell you how many work units you did, right? Accounting for the electricity cost is your problem, but they should provide the details of

        • Re: (Score:3, Informative)

          by vux984 (928602)
          Accounting for the electricity cost is your problem, but they should provide the details of who you donated it to that is needed for tax purposes.

          They would need to be registered charity though, for taxes. You can't just say you donated money to X and call it a day.
          • Re: (Score:3, Insightful)

            by evanbd (210358)
            They're a university. I'm sure they have that taken care of.
            • Re: (Score:3, Interesting)

              by vux984 (928602)
              They're a university. I'm sure they have that taken care of.

              Good point.

              Be interesting to see someone try and claim it though. I wonder if the IRS would agree to its validity.

              Probably help if they provided you with a proper receipt of some sort, which they don't.
              And I don't think it'll help non-americans even if they did, unless they were registered as a chairty in other countries as well.

              • by evanbd (210358)
                Exactly, some sort of receipt would be a good thing. But it shouldn't be very hard for them to provide enough of one, since they don't have to manage the conversion to dollars themselves.
              • They would need to provide a "receipt" showing work units contributed and what platform contributed it. You would then fill in the electricity cost in your area, and compute the "in kind" donation. All of this is moot though unless they're a 501(c)(3) charity.
      • by Frangible (881728)
        FYI:

        Old PS3s (90nm):
        Folding@Home with visuals: 215 watts.
        Folding@Home screen saver: 185 watts.


        New PS3s (65nm):
        Running Folding @ home 157

        Considering the GPU is still 90nm, that 157 figure should drop to ~127 watts when the screen saver kicks in.

        Typical energy costs are also more like $.10/kWh.

        127W x 24h/d x 365d/y = 1112520 Watt-hours/y or 1113 kWh/y
        at $.10/kWh that actually costs moar like: $111/y.
        Or if for some reason you're paying $12/kWh, that's still only $134, less than half of you
        • by Frangible (881728)
          update: I just measured my 65nm PS3 with a watt meter. Results:

          F@H with visuals (map + protein thumbnail): 148 watts
          F@H with screensaver: 131 watts

          Oddly the screensaver did not drop it by the same number of watts despite the fact the GPU is the same die size as on older PS3s; in fact, it was only 57% the decrease experienced on the 90nm PS3. I am uncertain as to why this may be.

          Regardless, 131 watts is still only ~$120/year for me, and that's certainly manageable for the sheer amount of work and fol

        • Re: (Score:3, Informative)

          by vux984 (928602)
          Your right I was off on the original data, by about $70 bucks, because I grabbed the wrong number. Early reports on ps3 f@h rated it at 280-300W, but this was corrected down to 190-220W thereafter. My mistake there and I've posted that I was wrong already elsewhere in this thread.

          As for the price of electricity, and your assertion that its 10c? vs 12c? Now were just playing statistics. I could justify mine by noting that prices are generally higher in Europe and Japan for electricity. (Its the equivalent of
      • I highly doubt the validity of the figures you quote for folding @ home's cost. This has been a perennial topic on /.

        But even if you were right, I'd say one cannot (for example) complain about the government not funding NASA and at the same time not run Folding @ home for economical reasons. Both are great science and both are worthy of (at least) a modest amount of investment.
        • by vux984 (928602)
          I highly doubt the validity of the figures you quote for folding @ home's cost. This has been a perennial topic on /.

          You are right. My figures were off. The PS3 watt rating for the 90nm version is 190-220 not 280. The 65nm version is more efficient at around 157.

          That brings the cost down to $188 and $137 respectively. But it bears mentioning that the .12c price I quoted was on the low end. It ranges from ~.05 to ~.24 in the US. In a state like new york its .17 so even with the corrected numbers its $356 an
    • It doesn't count as "examined" until we can see the source.
  • by Anonymous Coward on Wednesday April 16, 2008 @04:41PM (#23095414)
    I've been doing Folding @ Home for most of my adult life. I fold shirts, pants, underwear, etc. etc.
    • Re: (Score:2, Interesting)

      by cpricejones (950353)
      You can also visualize protein folding at home by going to www.pdb.org. The Protein Data Bank website has lots of cool structures to download, from small proteins up to large RNA-protein complexes like the ribosome (http://www.pdb.org/pdb/explore.do?structureId=2J00), which is one of the more remarkable achievements in structural biology. (Note that you may need a stronger graphics card to actually look at and rotate the whole ribosome as it is 64,000 atoms.) To actually look at these structures, you can us
  • Why visualize it? It's boring, and doesn't it use precious CPU/GPU power?
  • by Grokmoo (1180039) on Wednesday April 16, 2008 @04:46PM (#23095484)
    From the article:

    ET: Whenever someone hears about GPU-accelerated FAH, their first question is why there is no client with support for Nvidia cards. In the past it was said that it had more to do with Nvidia's drivers. Now that the core doesn't use DirectX, couldn't a GPU client use Nvidia's CUDA? Is there any work going on there, and if not, why not? Dr. Pande: We are interested in CUDA and are investigating how well FAH on CUDA would work.
    I am awaiting this with some serious excitement. Getting Folding@home working on Nvidia GPUs would definitely add a lot of computing power into the mix. This is especially true now, as it seems that the current crop of high end GPUs seems to favor Nvidia.

    From the benchmarks I have seen, it seems that there are currently no games that can effectively utilize, for example, 2 9800 GX2s. If Folding@home releases an Nvidia client, those people who have plunked $1000 into graphics cards may finally be able to put them to use!
  • Ati Only (Score:4, Informative)

    by Fross (83754) on Wednesday April 16, 2008 @04:48PM (#23095502) Homepage
    From TFA, interestingly this bypasses DirectX and interfaces with the card directly (I guess you'd want to, to throw maths at it instead of vertices)

    However it only runs on R600-based Ati cards right now. It also requires .Net framework. They do say they're "investigating" an nVidia version, but that sounds a while away.

    Interestingly also, it claims to parallelize processing the atoms, so it must use the individual stream processors on the graphics card directly.
    • by Trogre (513942)
      ... so no Linux version for a while then? A pity, since the OSS radeon driver has really improved in the past couple of years.

      I guess my cluster will sit there with just its CPUs crunching numbers and its GPUs idle for a while longer.

  • by relikx (1266746)
    So let me get this straight, you keep your computer running for long periods of time. The goal is "to understand protein...misfolding"

    Sounds like Pornography@home to me...
  • What about folding@work? I have access to 150 computers, most of which stay on 24/5, and do nothing 16/5 (okay, maybe 20/5).

    I, of course, would have to get the okay to do this, but I am not even sure I would want to...

    Has anyone done this? How did you go about it? What concerns are there (security, reliability)?
    • by Gewalt (1200451) on Wednesday April 16, 2008 @04:59PM (#23095642)
      Better check up on power consumption there, factor in 124 hours a week at 240 watts (conservative) x150 machines. Take a look at how much money your company is "donating" in raw power consumption, then triple it, cause for every watt of heat dissipated, it takes 2 watts of AC to remove it. See if your CEO approves of that donation to FAH that he can't even write off for tax purposes. (no receipt)
      • by iamhigh (1252742) *
        you should check out the Kill a Watt [p3international.com]. It tells you just how much power something draws.

        Just because you have a 240 watt PS, doesn't mean you pull 240 constantly. In fact with drives and monitors off, you might be pulling 75. At least for most of our common computers.

        I am environmentally aware, but I did the calculation and 16 hours of a computer running is less than 5 minutes of a $40,000 PHB's time. So the attempt to enforce the policy of shutting down computers nightly doesn't add up to the exec
        • by Gewalt (1200451) on Wednesday April 16, 2008 @05:20PM (#23095834)
          There's a HUGE difference between an idle computers power consumption and ones whose CPU and RAM throughput are being taxed to the limit by a process like folding. The 240 watts I mentioned is just the CPU, northbridge, RAM, and internal heat evacuation. Drives and monitor are completely irrelevant.

          You execs are right to dismiss the notion of shutting down a computer thats idle. It's NOT consuming much. However, when that same computer is crunching foldings numbers for it.... THAT is a huge cost.
          • by Klinky (636952)
            I don't know what computer you're running, but my Core2 running @ 2Ghz(speedstep) idles around 134W. When I run my 8800GT full bore and my CPU ramps to 3Ghz I suck up about 212W. I would find it surprising if your commodity office PC used more than 100watts @ %100 CPU. The gap between idle & full speed probably isn't that much either as a lot of computers are not setup to use SpeedStep or Cool N' Quiet by default.
            • by Gewalt (1200451)
              2.8ghz desktop P4 uses 130w at 100% and ~20w at 1%. Although the C2D is rapidly replacing P4s, there are still quite a few of them in the workplace. The earlier P4s actually consumed more power, and Rambus ram consumed about 3x the power as "regular" ram.

              BTW- you have a terribly inefficient computer if its sucking down 134w when idle... typical of an office computer is around 65w. Unless of course, you're talking about a gaming machine... (just realized the vid card you ref'd... d'oh!)
      • by iamhigh (1252742) *
        I re-read your comment... I was way off on what you were saying. Sorry for that.

        So is the program that intensive? Will it really pull that much power?
      • by kdkirmse (801423)
        I believe you have your ratio wrong. Modern AC units will move at least 3W of heat using 1W of power.
      • by specific (963862)
        Luckily, I live & work from a building that includes power in the rent. 8)
      • Just go to the statistic pages on WorldCommunityGrid.org. You will see that the top 4 contributors have over 5 million results. A thousand results per cpu year is a good rule of thumb so 5 million results is about 5000 cpu years. At $300 per cpu year that amounts to 1.5 million dollars so there are large corporations that are willing to donate that much time and effort.
      • by Trogre (513942)
        Sorry? 2 Watts to remove every watt of heat? Surely your A/C units can do better than a COP of 0.5

    • by scubamage (727538)
      At my high school all computers had folding@home installed. Usually there is little strife from layer-8 types because the machines aren't doing anything during that time anyways, and a decent marketing department can spin it as some awesome PR. At the very least if you have install rights on your machine you can run the client on your workstation.
    • I used it to test out a new server once just to make sure it was stable, after having crashing/network card issues during installation of Windows. Being the IT manager (okay, so the only IT staffer, hehe) I didn't really need to get the go ahead to do that though :P I agree that for that many computers, the power consumption costs really would be too many. For your own workstation it would probably be fine. I've noticed one engineer at work had his screensaver set to do climate change calculations or someth
    • by Verteiron (224042)
      An interesting problem with doing this comes from an unexpected source: noise. Most of the PCs I've set up here have fans that spin faster the warmer the machine gets. When the CPU is pegged at 100% for more than about 5 minutes, the workstation sounds like a Harrier lifting off. Now imagine an entire building full of these...
  • by denis-The-menace (471988) on Wednesday April 16, 2008 @05:17PM (#23095798)
    "to understand protein folding, misfolding, and related diseases."

    FYI: This means Prions related diseases => Mad cow disease
    • Re: (Score:2, Offtopic)

      by bradbury (33372)
      Actually this is a grossly incomplete statement. Due to the fact that the process of Non-homologous End Joining (NHEJ) repair of DNA double strand breaks involves the exonuclease (DNA end-eating) proteins WRN and DCLRE1C (Artemis), the repair of double strand breaks corrupts the genome via microdeletions. Microdeletions can result in frameshift mutations which can of course result in protein mis-folding. The accumulation of these frameshift mutations and misfolded proteins over the lifespan of cells has
  • by Viking Coder (102287) on Wednesday April 16, 2008 @05:27PM (#23095902)
    So, here's my thought - before someone sends an email, they contact Folding@Home, identify themselves, say who they want to send an email to, and the contents of the email. F@H gives them a work unit. When they complete it, F@H signs their email. Your email client can filter emails based on how many work units the sender did to send it to you. If someone really wants your attention, they'll process for a day or two. If it's a casual email, one work unit will do. Maybe even a fraction of a work unit.

    That way, if you read spam, at least you know that you contributed to F@H. If you want less spam, you turn up your threshold for how many work units the sender has to do.
    • Uh. I think I'd rather use the phone. And that's really saying something. Usually when someone needs my attention, they need it very soon.. talking about work situations here of course, for personal email your method would be crap to middling.
      • So make it fractional, and allow people to build credit. Have F@H give out credits at about the rate of about 60 per hour of work against a Work Unit. The entire concept, duh, is that you trust people you know, and you demand credit from people you don't know. If that doesn't work for at least 50% of your work and 95% of your personal email, then you don't trust many people... So ask for more credits before you'll unfilter spam.
        • Limping along with the current system of email is just a bad idea anyway, it needs to be secure from the ground up, not just tagged on after. Spammers can generate a whole lot of credit with a botnet as well, so the only person that gets inconvenienced here are those that are trying to send legitimate mails. They also have to contribute processor time and energy to a cause that they might not be wanting to support (if they contribute to another charity or whatever)..
          • So, allow multiple crediting agencies. The same way I can pick a Star Wars stamp at the Post Office.

            If they've got a botnet, and I can force them to use it to help F@H, I see that as a good thing.

            And as I said in another post, if someone sends a legitimate email, I think it makes sense that you could return their credit when you say, "Ah yes, this isn't spam."
    • by evanbd (210358)
      All the joys of hashcash, with none of the headaches of writing a cross-platform solution! And there's a third party involved, too!
      • Re: (Score:3, Informative)

        by Viking Coder (102287)
        The stupid part about hashcash is that no useful work is done. I'm proposing an idea that would at the very least get something useful accomplished.

        There are all sorts of third parties involved in sending email. I'm not proposing a solution for everyone - I'm suggesting one possibility.
    • Re: (Score:3, Insightful)

      Yeah, because Lord knows the spammers don't have spare CPU capacity at hand on all the hijacked machines they control.

      It's worth a shot at thinking outside the box, but they have the CPU cycles and can likely hack past any kind of attempt to node lock the work units.

      I suppose a minor benefit would be that some kind of work gets done before a spam message was sent out, but there's got to be a way to get past that requirement -- F@H is based on a measure of trust (and some cross-validation) that parti
      • So, you increase the amount of cross-validation. I kind of fail to see the problem here. At the very least, you're contributing to F@H. That's a good thing.
  • How about we just use this huge processing network to emulate actual neurons and link them together as close to a brain as we can. Then we can see what happens..

    i know it takes billions of neurons to do anything, but with all this extra power laying around we might just have enough to do it.

One man's constant is another man's variable. -- A.J. Perlis

Working...