Slashdot Mirror
Distributed Computing on Next Gen Consoles
anonymous lion writes "Wired has a story on the need for Xbox 360 and PlayStation 3 to support distributed computing with a non-gaming purpose. The article goes on to discuss SETI@home, distributed.net, and Folding@Home." From the article: "The next generation of console gaming is going to see a huge increase in machine performance and overall computing power. Already planned for both the Xbox 360 and the PlayStation 3 are multiple 3.2-GHz PowerPC processors capable of handling advanced gaming and graphics simulations, along with out-of-the-box internet capabilities such as Xbox Live Silver. With all that horsepower in a machine that is used for only a fraction of a day, we should offer gamers a chance to put these unused resources to good use."
How about switching the thing off?
Its not that a game console is something like a desktop pc, running the whole day just to be quickly accessable....
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
And most people won't offer to have their console used for Seti or folding or whatever. Something that's needed more then horsepower, is the willingness to bother with it. And that will stop too many of these things from being overly popular.
It's great to help cure cancer, but not if it causes the number of polygons on Lara Croft's breasts to drop.
Also, will users have a choice concerning whether to so use their consoles' spare cycles, or will it happen without their concent or even overt knowledge? Will they be able to decide which project gets the use of their machine's time? And what if someone comes up with an entertainment use for those cycles...?
Sure,
the average consumer LOVES to waste power and bandwidth to search for aliens. Folding, Seti & others are good projects, but if Wired thinks the average console owner wants his console to suck power, bandwidth, and make huge fan noise while not doing something with it,they may be seriously mistaken.
I'm sure the same people that run Linux on their XBOX will run folding on their console, but not the majority of users, even if the console ships with that functionality.
I'm fed up with the idea that sharing is caring. I don't like to share. I don't want anyone using my bandwidth, my CPU-cycles, my harddrive or my bathroom. It's not that I don't have a high bandwidth connection or several idle CPU's laying around, it's just that I don't like the idea of giving when all I want is recieving (i.e. torrent) I say we put an end to this hippiecomu P2P and other distributed services once and for all.
I think someone has confused something they would like with an actual requirement. I can just see all the parents lining up to subscribe to this 'need' because they really want to use their jumped up electricity bills to help search for extratrestrial life signs
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
All of the new console makers are going to be losing mega cash for each console sold, so why would they make any incentive for anyone to buy their consoles and use them as computers? The manufacturers lose money on the console and lose any possible revenue from game sales.
.01% of the uber-nerd population is insanity.
If the console manufacturers provide software that somehow taps the raw horsepower of the new consoles what would stop organizations, legal or not, from buying large quantities of game systems just to make a supercomputer for very cheap? Fuck that.
If I had not preordered my PS2 a year in advance I would have had to wait NINE months to be able to get one in the states. The demand for the new systems is going to be even greater. The last thing consumers need to hear is that there is a shortage of their favorite game system because Nerd University bought 10,000 systems for their new supercomputer project.
Shared computing is all fine and good for PC/Mac users, but honestly, for a manufacturer to open the floodgates of their OS to satisfy the wants of
"Jeremy, you need to get to an internet cafe and cut and paste some appropriate sentiments about me from the world wide
At least with current platforms architectures. The author seems to do plenty of research on current distributed computing projects, but does none on how the consoles perform.
I know that SETI@home has been ported and tested at least on the XBox, and it performs miserably. These console gaming systems are designed to play games, not do radio signal analysis or other scientific calculation. For example, there's little need for fast memory writing when you're mostly reading textures from RAM, but there's an extreme need when you do millions of in-place Fourier transforms. Unless Microsoft and Sony change their architectures for some inexplicable reason, I can't imagine future architectures would perform much better.
This article smacks of ignorance on the part of the author, who clearly did no research into the actual performance of consoles in regard to standard scientific computing.
Since distributed computing projects crank your CPU to 100%, there's definitely an associated energy and environmental cost to running that stuff. This will become increasingly true in the future, with the increasing prevelence of technologies like Intel's "SpeedStep" or AMD's "Cool And Quiet" that allow CPU clockspeeds to dynamically vary the clockspeed and power consumption of a processor. That will only increase the difference in power consumption between a CPU at rest and a CPU that's pegged at 100% crunching SETI units.
Distributed computing advocates always seem to neglect this. They think that all those unused CPU cycles are a vast, untapped resource just waiting to accomplish fabulous things. Well, as a guy who used to have a few boxes crunching RC5-64 for Distributed.net, I can tell you that it's not a free resource when you're the one paying the electric bill.
Joe Consumer isn't necessarily going to think this technology is a great idea when he realizes that he's paying an extra $10 a month on his electricity bill for the "privilege" of crunching numbers for some dubious cause.
And, let's face it. Not all distributed projects are dubious, but many are. The fundamental problem is that a lot of compute-intensive projects simply aren't embarassingly parallel like SETI or RC5-64. And a lot of other parallelizable applications require access to huge datasets that make them unsuitable for distributed work. For example, 3D rendering can be parallelized pretty well... but the datasets are huge. For your CPU to render a single frame of Pixar's latest movie, it would need access to anywhere from hundreds of MB to several GB of texture and geometry data. A lot of scientific applications are similarly constrained.
OtakuBooty.com: Smart, funny, sexy nerds.
Well, it may not be much of an issue now, but this is quickly changing.
Both Microsoft and Sony are playing with the idea that these game consoles will do more than merely play games. If it also has DVR functionality, advanced DVD capabilities, etc., then the day will soon arrive where people DO leave them on 24x7.
I have a TiVo, which is just a special-purpose computer. I wouldn't mind at all if it had a "power down" mode that would run a grid application such as trying to help cure cancer while it's not recording anything and I'm not watching anything.
And for what it's worth, both my Xbox and PS2 stay on 24x7. When I'm done playing a game, I usually just switch the input on the tv, not actually turn the console off. Am I alone in this?
If Martz wants to use the video game consoles and electricity of people to do his calculations, let him give the people something they want in return, like free games.
That would probably be enough to motivate a lot more people to turn their machines over to SETI.
The idea that people are going to let their machine run their crunching away, for free, for no benefit, is pretty stupid. The first distributed computing project to offer any sort of tschocke is likely to become more help.
http://www.thebricktestament.com/the_law/when_to_
What if some company would find a smart way to pay the users for the use of their CPU cycles?
Anything like what you described (or any compensation for your CPU cycles) is unlikely to ever happen. Reason? Most of the organizations asking for your CPU cycles are either too poor or too cheap to give you anything in return. They can't even afford to pay for the power usage that you incur, let alone put anything towards your hardware.
And for what it would cost to create and maintain a MMOG like what you're talking about, at least one people would be interested in playing, they could just buy an assload of computers (think $100 to $200 a pop barebones systems) and plug them in.
Not that it isn't a cool idea, just not feasible. You have to see the organizations asking you to run this software for what they really are...beggers. Not that that's necessarily a bad thing, as long as you realize they will probably never have anything to offer you (other than the warm and fuzzy feeling of geekiness).
(1) All of the distributed applications that you mention release the results of their research as public scientific publications. Any companies can use the results, but so can anyone else. Subscription to the journals is all that costs money, but generally free "e-prints" are available. All of the distributed applications that you mention are non-profit.
(2) Even if they were patenting the results (which they aren't -- see 1) it is better to have the patented result that one has to pay for than to have nothing. If I have breast cancer, I would rather pay $1000 for a test than be unable to get a test because no company wanted to invest in it.
As a side rant (somewhat related to (2)), you say patents are inhibiting progress. But without the financial incentive that the breast cancer patent generated, the medicine would never have been developed. I'm sorry that so many people only work out of greed, but that's reality at the moment. And it actually works pretty well.
Perhaps IBM doesn't just want to sell chips to these people. Perhaps it has a reason for selling the PC division to Lenovo. Perhaps it sees an opportunity to create a business architecture in which the virtual business world runs on the server farm, while the graphics and sound capability of the very cheap clients delivers a superior user experience that makes users happy not to have a "PC" on their desk. Meanwhile the data mining and compute-intensive activities are farmed out to those clients while they aren't being used. Fault tolerant. Cheap to extend. And round objects to Microsoft.
Panurge has posted for the last time. Thanks for the positive moderations.
the article talks about distributed computing on large scale which is not very feasible for all the above mentioned reasons like
a) companies wouldn't spend money on building this into the console
b) most consumers cannot be bothered
There will be people who'd be interested though. I'd try it if I had a ps3... However with more and more use of clustered smaller machines in place of large supercomputers, clusters of consoles have been built in unis and research labs (for example here or here. There are a few advantages to using consoles:
a) they are cheaper
b) they are small form factor
c) they have hardware optimised for computation (at least ps2 does and ps3 will).
Sony had released linux on ps2 and word is they will be releasing linux for ps3 with extensions for the Cell's SPUs. Once ps3 has a fully featured OS any scientific app can be ported and modified to run on it. Now M$ on the other hand, well, I don't see them releasing any OS for XBOX 2*Pi but maybe the xbox linux crowd will take care of that.
This is exactly my point. Individually the processor may perform well, but when it's placed in the actual system, perform will undoubtedly drop. Right now, I'm doing performance tests on FFTs performed on GPUs (graphics cards). Theoretically, these should perform at the same "incredible" speed as the Cell processor (10 Gflops or better), but in reality bandwidth and cache constricts performance to half a Gflop.
You'll also note they say nothing about the system they tested on. Was it a PS3? I bet you it wasn't. Let's wait and see the actual performance of the PS3 before we get excited from in house tests done on an unspecified system by a company that is eager to boost impressions of its new chip.
1. I bought my first Mac in February. Now it seems PPCs are not in the Mac future.
2. I run ClimatePrediction.net on my Mac and Linux x86 systems. The program is huge, comes from a mainframe environment, and is married to an INTEL compiler. The PPC version is, needless to say, not very fast. Single work units can take months to complete.
The other projects in the article would be on my plate, too, if they compared with my concern for climate change.
Something I've thought about a few times, but done no research on whatsoever, is the possibility of distributed computing on mobile phones. They are getting more and more powerful, more and more ubiquitous, and they are connected to a network. And they're kept on all day with their processors idle for the most part. We could use this computing horsepower for somthing that will benefit all of society, such as processing marketing data.
Ok, I stand corrected here. I have seen distributed computing come up where things were not going to be released back to the public though. Most universities, including the afore mentioned Stanford, are doing research with corporations who get to monopolize the results when something useful comes out of them (and taxpayers subsidize university research departments). Although this article doesn't indicate that one way or the other. It does give a link to the project but I don't really want my organic/bio-chemistry to come flashing back at me.
Point 2:
The company that I was talking about in Utah didn't patent a drug, they patented the gene! It just grants them an exclusive right to persue things related to that gene. They haven't come up with anything more than a test to see if one has the gene. They have used that patent to stop many other companies and universities from persuing drugs and gene therapies related to that gene (source: 60 Minutes). I can see granting the patent on the test, but not the gene! In my opinion that is just wrong.
Restore America: Dr. Ron Paul for President!
The Cell processor (PS3) is made for those applications. At the Power.org convention in Barcelona, IBM presented a programming example of large FFTs on Cell. It turned out, that large FFT calculations are about 100 times faster than on a Xeon 3.2 GHz processor.
Keep in mind, that this presentation was held in front of super computer professionals and its not that easy to trick them.
"With all that horsepower in a machine that is used for only a fraction of a day, we should offer gamers a chance to put these unused resources to good use."
Now the Captain is wondering how many of us actually leave our consoles on when not in use? Show of hands... Now! Hmmm, not too many. Now how many of you would actually like to pay extra in electric bills to do it? Ouch. Even less. And finally, how many are going to mod their PS3 and actually downloard the app to make it happen? That leaves just about... Nobody.
You need a FREE iPod Nano
Apart from the fact that during gaming itself the Distributed Computing program should be (and probably will, as they do now) in the background doing nothing, and that they shouldn't run 100% but rather a bit lower when the console is idle, I could definitely see this happen if the game manufacturers give bonus levels, more bullets, stronger armour or whatever is good, nice and fun in a game. For the game manufacturers it will generate a ton of free publicity when their game helps curing cancer. The gamers get more fun. And if you don't want that, you turn it off. Now how hard is this?
Bert
Bonus sllogan: Save a live and you get an additional live.
MS doesn't have PowerPC CPUs, those cores are in-order PPC-based chips for embedded systems (same goes for the Cell's all-purpose core). They'd perform awfully in PCs. Chris Hecker claimed the chip is 3-10 times slower than a comparably clocked PPC in the Burning Down The House session at GDC. A P4 at a similar clock speed would kick the shit along with the intestines out of that "PPE".
It's actually not known what kind of chip Nintendo will use, not the clock speed, not the features, nothing except the codename "Broadway".
Justice is the sheep getting arrested while an impartial judge declares the vote void.
Do the "longer play sessions" of a game console continue through the night? This isn't the era of the NES, where 5-hour games didn't have a save feature *cough*Super Mario 3*cough* and players would leave the console on pause overnight. Besides, a TV can be used for only one thing at a time, and if it's not playing games through a console or playing DVDs through a console or other DVD player, it's either off, showing cable TV, or showing satellite TV.