Forgot your password?
typodupeerror
The Internet Programming IT Technology

Distributed Computing "Advances" 160

Posted by Hemos
from the yet-more-clock-cycles dept.
Quirk writes "NewScientist is reporting on..."Software to be launched in January will let PC users run as many "distributed computing" projects as they like. The program will let PC users search for aliens, help predict climate change and perform advanced biological research - all at the same time."'It is called the Berkeley Open Infrastructure for Network Computing (BOINC). BOINC acts like a software platform that can run a number of screen-saver style applications on top of the PC's own operating system.'"
This discussion has been archived. No new comments can be posted.

Distributed Computing "Advances"

Comments Filter:
  • pretty sweet (Score:3, Informative)

    by lotas (177970) on Monday December 22, 2003 @09:51AM (#7785532) Homepage Journal
    im already running boinc on a few of the machines at home and work and it works cool. i especially like the built in queing and multi processor support.
  • About Time! (Score:5, Funny)

    by Joe U (443617) on Monday December 22, 2003 @09:51AM (#7785533) Homepage Journal
    Finally, a source for my advanced alien biological climate change program!
  • by Gothic_Walrus (692125) on Monday December 22, 2003 @09:51AM (#7785536) Journal
    Does this mean that now we'll be mapping het the genome of aliens with AIDS?
  • by questamor (653018) on Monday December 22, 2003 @09:51AM (#7785541)
    The first project underway in BOINC is to have everybody's machine submit news about BOINC to slashdot, which is so far happening succesfully. This is the first dupe of many.
  • by Rosco P. Coltrane (209368) on Monday December 22, 2003 @09:53AM (#7785553)
    The first and easily the best known is SETI@home, which since 1999 has enlisted half a million people to analyse data from the Arecibo radio telescope in Puerto Rico, looking for signs of alien life.

    Better than Seti@home and BOINC: Yeti@home [phobe.com].
  • by Palverone (166646) on Monday December 22, 2003 @09:53AM (#7785554) Journal
    Even though you *can* do multiple projects at one time, you have to run seperate applications (if I'm correct) so this would be a good integration into one application that handles multiple projects and allows your machine to be used more efficiently.
    • I belive the great achievement will come when it supports multi processors. I don't know much about clustering, but is it possible to run a program like that on a Linux cluster for example ? And how about those Intel Hyper Threading processors ?
  • by sosegumu (696957) on Monday December 22, 2003 @09:57AM (#7785573)
    Have you ever thought that the internet is just one giant 'distributed computing' effort to find pr0n?

  • Who is Benefiting? (Score:4, Insightful)

    by Famatra (669740) on Monday December 22, 2003 @09:57AM (#7785575) Journal
    I was interested in the folding protein project, but are the results open to the public (like the human geneome project) free of charge, or will someone making a buck off *my* computing power?

    With all the distributed computing projects out there be sure to read the fine print, if your going to use your computer for a project make sure its helping everyone instead of a few corporations make $.
    • by Anonymous Coward on Monday December 22, 2003 @10:05AM (#7785615)
      From the Folding@home website FAQ:

      "Who "owns" the results? What will happen to them?

      Unlike other distributed computing projects, Folding@home is run by an academic institution (specifically the Pande Group, at Stanford University's Chemistry Department), which is a nonprofit institution dedicated to science research and education. We will not sell the data or make any money off of it.
      Moreover, we will make the data available for others to use. In particular, the results from Folding@home will be made available on several levels. Most importantly, analysis of the simulations will be submitted to scientific journals for publication, and these journal articles will be posted on the web page after publication. Next, after publication of these scientific articles which analyze the data, the raw data of the folding runs will be available for everyone, including other researchers, here on this web site."

      http://www.stanford.edu/group/pandegroup/folding /
    • Yeah...and it could be worse. If Real Player were running a 'distributed computing' project, they would probably charge you for them to use your computer.

  • Double work (Score:3, Interesting)

    by enodev (692876) on Monday December 22, 2003 @09:58AM (#7785577) Homepage

    "Keeping track of how much work everybody has done is one of the prime motivations," says Anderson. BOINC checks this by farming out each problem twice and comparing the results. "If the answers are different we have to assume that one of those parties may have cheated," he says.

    So the whole work has to be done twice for the sake of correctness. I think they should introduce some trusted user mode, let's say, so that results from users who have invested a certain amount of cpu time should be trusted or at least not every received result double checked. Just every n'th packet or so and if it's invalid they have to recheck all unchecked packets. I guess this would reduce double work a lot as there is normaly only a minority of users who's trying to cheat.
    Does this sound sane?
    • Re:Double work (Score:5, Insightful)

      by jkcity (577735) on Monday December 22, 2003 @10:23AM (#7785715) Homepage
      Double results and checking also helps to capture random errors i would guess as well though, not just cheating.
    • I think they should introduce some trusted user mode, let's say, so that results from users who have invested a certain amount of cpu time should be trusted [...]

      How do you know how much CPU time a user invested? That's also information submitted by the user that you cannot trust.

      Unless of course the "trusted user mode" involves requiring the user to run the software on some sort of "trusted" computer, Microsoft Palladium style. In that case you might be able to save 50% computing time, but I suppose at

  • by Anonymous Coward on Monday December 22, 2003 @10:00AM (#7785586)
    I've always had some mild reservations about running the closed-source SETI code, but convinced myself it wasn't an unreasonable exposure. A meta-app that exists to download yet more closed-source code without telling me... nope, that's over the line. Sorry, lil' green guy, but this is too much to ask.

    (signed) a top 1% setiathome client.
    • A meta-app that exists to download yet more closed-source code without telling me...

      Sounds like Windows Update on the automatic setting. :^)

    • by SETIGuy (33768) on Monday December 22, 2003 @01:25PM (#7787185) Homepage
      A meta-app that exists to download yet more closed-source code without telling me... nope, that's over the line.

      The SETI@home (under boinc) source code [berkeley.edu] is available under the GPL. The AstroPulse code should be available shortly. Yes, now you can see how bad my code really is.

      What you won't get with the code is our code signing key (which is under lock and key on an isolated machine) or the ability to distribe your version from our servers, but you are welcome to compile versions for use on your machines and/or distribute your own versions. We won't guarantee to anyone that your version doesn't erase harddrives or distribute child porn, though.

  • by BillGodfrey (127667) on Monday December 22, 2003 @10:01AM (#7785590) Homepage
    If you didn't read it first time, here it is again...

    My Primer on building a distributed computing project. [bacchae.co.uk]

    (It still needs updating.)

  • BOINK (Score:5, Funny)

    by Dylan2000 (592069) on Monday December 22, 2003 @10:03AM (#7785598) Homepage
    Wouldn't it make more sense if they'd chosen a last word beginning with a K?

    Boinking aliens and cancer with my computer? Sign me up!
    • Re:BOINK (Score:2, Funny)

      by sharkey (16670)
      Wouldn't it make more sense if they'd chosen a last word beginning with a K?

      I'm looking forward to the Berkeley Open Infrastructure for Networking GNOME (BOING). Maybe they could get Berke Breathed to design the mascots for it.

  • by Bowie J. Poag (16898) on Monday December 22, 2003 @10:03AM (#7785599) Homepage


    Using "quotation marks" in the "wrong places" makes everything you "say" seem "suspicious".. Like you're trying to "pull one over" on the "reader" by insinuating theres a double "meaning" to the "word" in "quotes"..

    Hate to be a grammar Nazi, but, the the whole quotation mark thing is a pet peeve. :)

    Cheers,
    • Using all those quotes reminds me of Chris Farley's character Bennett Brauer (spelling?) on SNL.

    • Using "quotation marks" in the "wrong places" makes everything you "say" seem "suspicious".. Like you're trying to "pull one over" on the "reader" by insinuating theres a double "meaning" to the "word" in "quotes".

      You're absolutely "right", nothing annoys "me" more than overuse of this "technique". I "literally" claw my eyes out everytime someone "misuses" quotes.

    • But in this case, it's correct, because the intention is to place suspicion on the word 'advances'.

      Maybe you ought to make a new year's resolution to reduce the number of things that bother you. It'd be better on the rest of us who are constantly reading your dumb pet peeves. Hate to be a grammar Nazi my ass, you love it.
  • YAPSFUAS (Score:4, Funny)

    by cyclist1200 (513080) on Monday December 22, 2003 @10:03AM (#7785601) Homepage
    Yet Another Project Suffering From Unfortunate Acronym Syndrome.
  • by Indy1 (99447) <spamtrap@fuckedregime.com> on Monday December 22, 2003 @10:03AM (#7785606) Homepage
    typical reporters fscked their facts in the story.

    qoute "The first and easily the best known is SETI@home, which since 1999 has enlisted half a million people to analyse data from the Arecibo radio telescope in Puerto Rico, looking for signs of alien life. [newscientist.com]"

    I believe distributed.net's [distributed.net] client was the first program of its type to download information from a remote server, use idle cpu cycles to calculate whatever, then resubmit it back to the central server. I ran distributed.net back in 98, more then a year before seti came out.
    • According to their site [berkeley.edu] the first release was on 6/8/98. Not sure if distributed.net was before that, but you weren't running it "more then a year before seti".

      They were way off on the user stats by nearly an order of magnitude. The statistics [berkeley.edu] page shows over 4,800,000 users.
    • by stevey (64018) on Monday December 22, 2003 @10:50AM (#7785941) Homepage

      I'm not sure that I can prove this, but I created a distributed client of sorts in 1997.

      It was a java applet which ostensibly did some cute" image animation, back when such things were new and fun to write.

      What it actually did was download from my server the latest value of PI and try to compute more digits. When the applet was destroyed it submitted its result to the server.

      It was fun watching the result get gradually longer and longer with no effort on my part just due to people who were interested in my webpages.

      Maybe it should have been advertised, but I took pleasure knowing what was going on ..

    • We're talking about network queueing systems here.

      General purpose queueing systems have been around a loong loong looooonngg time; 20, 30, 40 years. Distributed.net and SETI simply expanded the concept to include other people's computers. Hell, NASA produced a freely available and popular one in the 80s called NQS which is still available.

      I have to laugh at the thought that all this "Grid" and distributed stuff is new.

    • the GIMPS project has been around longer than both of them, and unless I'm mistaken, is the longest running DC project currently available.
      • Yes, this is absolutely true. GIMP was around when d.net started and they're still going strong today. There was also Rocke Verser's DESCHALL group which had a head start on distributed.net by a couple months, but they shut down when they completed the RSA Labs DES challenge.

        Seti came well over a year later.

        For d.net, at least, our first assigned block was in early March 1997.

        http://www.distributed.net/history.html.en
    • I once got hit by a worm that installed a D.Net client in the background. I didn't notice it for two or three months.

      I damn near fell over laughing when I found it. :)
    • Distributed computing toolkits go back AT LEAST to 1973 and before with DCS. It's not like the d.net client was the first one I or someone else ever wrote. That's why it only took us a couple weeks from when Genx pulled the plug to write what whould be known as the d.net client.

      But yes, SETI does always claim to be the first.

      They are claiming to be the first "multi-project" client too, but you all remember picking between DES and RC5 I'm sure :) Folding@home and others are multi-project too, and that was
  • by AtariAmarok (451306) on Monday December 22, 2003 @10:06AM (#7785618)
    Judge: "What do you have to say about the virus you created, young man?"

    Virii writer: "It wasn't a virus, your honor. It was really a non-permission-based propagation model for a distributed computing application that involved producing the results of decreased uptime and further propagation of the non-permission-based distributed application."
  • by xaoslaad (590527) on Monday December 22, 2003 @10:08AM (#7785634)
    http://boinc.berkeley.edu/

    I didn't see it in the story either. Pardon me please if I'm just blind/illiterate
  • Skeptical (Score:5, Funny)

    by Root Down (208740) on Monday December 22, 2003 @10:10AM (#7785642) Homepage
    ... because we all know that no really good concept in computing has ever come out of Berkeley. ;)
    • Re:Skeptical (Score:2, Insightful)

      by RicktheBrick (588466)
      This only works if one is using one's computer for personal reasons and contributing the wasted cyles to the cause. If one leaves the computer on only to do the calucations than one is paying more for the electricity to run the program than the calculations are worth. I refer to the $5 million dollar supercomputer at Virginia Tech. This computer can do 8 trillion calculations a second. Now how many pc would it take to equal that and than caculate the cost of electricity for that amount of pc's. Now the
  • What I'd like to see (Score:5, Interesting)

    by elliotj (519297) <slashdot@elliotj ... m ['nso' in gap]> on Monday December 22, 2003 @10:12AM (#7785654) Homepage
    I'd like to see a distributed computing app that can be used to both do the work (like the current ones do), AND optionally have the ability to submit a task. This way you could have a world wide supercomputer that everybody would have a chance to employ. Very few people would probably use it, but it would be very interesting to see the ways in which different people put it to use.
    • I've been thinking about something like this all semester in my Distributed Computing class.

      What I'd really like to see is a system setup where you have a network of clients, any of whom can dispatch an agent across the system that consumes resources to accomplish some goal.

      Obviously there would have to be some sort of non-malicious code signing or sandboxing going on within the system, as well as forcing the agents to consume proportional resources (ie the more time/space/bandwidth you give to the syt
    • Add some sort of payment to the mix, and it might work. You get CCredits for doing computations and pay C(ompute)Credits to get something computed. Projects have to acquire CCredits for their computations by some kind of authority, commercial ones for real money, good cause / scientific ones for free. Clients can then use their CCredits for their own projects (a la your "submit a task"), or (if coming from a commercial project) cash them in or donate them to a good cause. At some kind of exchange the client
  • BOINC acts like a software platform that can run a number of screen-saver style applications on top of the PC's own operating system.
    Sounds like Internet Explorer to me... Can't get much more "distributed" than the virii hiway of MS Products! And the nice thing is you don't even have to trouble the user to install your clients...
  • by devnullkac (223246) on Monday December 22, 2003 @10:13AM (#7785661) Homepage

    Scientific progress goes 'BOINC'?


  • by Seek_1 (639070) on Monday December 22, 2003 @10:17AM (#7785673)
    This really isn't as good as you might think.

    Most distributed computing projects are distributed because they need massive amounts of CPU cycles. Running multiple projects on one machine isn't going to make the projects faster since the same amount of CPU cycles are now being divided up amongst the number of projects that you're running. Infact it'll actually be less because now the machine has to deal with the overhead of switching between project processes.

    On the other hand it might make sense if you were running a CPU-intensive project and a data-intensive project at the same time (ie projects that will maximize separate non-conflicting resources on the same machine..)

    My Folding@Home Team [stanford.edu]
    • Pre-emptive response to any RTFA posts.

      I realize that BOINC won't necessarily be running projects concurrently on one client machine, but the point remains the same. There is not a very significant advantage to having something like this. It's not like Stanford is running out of data for Folding@Home.. (which would be the only reason that you would need to switch projects..)
    • Running multiple projects on one machine isn't going to make the projects faster since the same amount of CPU cycles are now being divided up amongst the number of projects that you're running.

      The default BOINC operating mode on single processor machines is to alternate projects to balance work between projects.

      But that's not really the point. I'll assume you donate to charities. Do you only donate to one charity? Probably not, becase there is more than one worthy charity.

      I think that there is more

  • by Anonymous Coward on Monday December 22, 2003 @10:18AM (#7785683)
    Well....the processors in my computers are OWNED by me. I pay the electricity bills to operate them, and YOU want to use my processor time for FREE ?? I dont think so, pony up some cash or keep your distributed clients, thank you.
    • I suppose contributing to the collective good doesn't turn your crank then does it?
    • "Well....the processors in my computers are OWNED by me. I pay the electricity bills to operate them, and YOU want to use my processor time for FREE ?? I dont think so, pony up some cash or keep your distributed clients, thank you."

      I run SETI on my PC all the time. Its cheaper for me to leave the computer on at all times and use it to maintain a consistent temperature during the winter with it than to crank on than the apartment's electrical heater... :)

  • by TrekkieGod (627867) on Monday December 22, 2003 @10:21AM (#7785698) Homepage Journal
    Oh...that was the sound of a million auxiliary generators being turned on to counter the increased power needs of all these processors.
  • by Boricle (652297) on Monday December 22, 2003 @10:22AM (#7785704) Homepage
    I always thought that it was the

    Search For Terrestrial Intelligence [totl.net]

    I know I've been struggling... have you found any? Will this help?

  • Jorn Wittenberger's Askemos project may be intresting for you.
  • this is amazing....i can't believe ive never heard about this before....on slashdot....oh wait...
  • by Dan Crash (22904) on Monday December 22, 2003 @10:29AM (#7785754) Journal
    I know one of the reasons they created BOINC is that the current SETI@home clientbase is very rigid and can only process data from one telescope -- Aricebo. I also know that the commandline client is tons faster than the screensaver-based client. Is BOINC's flexiblity going to end up making BOINC clients slower than the current dedicated clients?
    • is that that the screensaver is only less efficient when displaying its graphs?? how does it compare to the command line versin once the screensaver switches to go to blank mode (thats a setting directly in the screensaver not just a power save of the monitor....)
      • how does it compare to the command line versin once the screensaver switches to go to blank mode

        I'm not sure what happens when the screensaver switches off. However, I do know that when using SETI@Home on a P4c (with HT) and the SETI window is brought to the front, the processor usage shoots up from 50% to 100%. Maybe this is due to a separate thread being spawned that can be used on the second logical processor. I'm not sure.

        This is one thing I would like about the new system, the ability to run mo

  • by AndroidCat (229562) on Monday December 22, 2003 @10:30AM (#7785757) Homepage
    I'm sure that spammers will be registering their distributed spam/DDoS zombies [theregister.co.uk] real soon. Why sneak the software onto machines when you can get people to sign up for it if you provide fancy ratings and team standings? Throw in some t-shirts and blue pills and they're gold!
  • by Anonymous Coward
    is there anything BOINC-ing can't do?
  • Compared to OGSA? (Score:2, Interesting)

    by jonasmit (560153)
    This may be great for a few high profile applications that users are willing to support. But the Globus Toolkit OGSA project has higher ambitions OGSA [globus.org] and arguably a better chance of making a difference in the next generation of the WWW.
  • graphics and Boinc (Score:3, Interesting)

    by The Lynxpro (657990) <lynxpro@gmai[ ]om ['l.c' in gap]> on Monday December 22, 2003 @11:02AM (#7786029)
    From my understanding, Boinc uses OpenGL to unload the screensaver graphics off the main processor's load and onto the graphics card GPU just like how Mac OS X accelerates its GUI graphics (or how Longhorn will do it with DirectX). Too bad Boinc can't uses the GPU like what was covered here on Slashdot under the BrookGPU project yesterday...

    • by SETIGuy (33768)
      Too bad Boinc can't uses the GPU like what was covered here on Slashdot under the BrookGPU project yesterday...

      Some people have expressed interest in getting BOINC to do that. It may happen.

      • "Some people have expressed interest in getting BOINC to do that. It may happen."

        Good, because I'm interested in seeing how many gigaflops my old Pentium 133 would produce with five (5) Voodoo1 cards filling up all the available PCI slots. My friends and I have plenty of spare old videocards to donate to such an endeavour... :)

  • I downloaded bonic in January 2003 after reading Prey [amazon.com] by Michael Crichton. You have to read a lot of documentation to get going because you must work within their framework. After fiddling for several hours I gave up, because I didn't think many people would bother to run my distributed "Hello, World" application. You see, each client computes the ASCII value for a character in the string, the server then reasembles them and prints it on the server. It greatly reduces the work required to display output on
  • SIC@HOME (Score:3, Funny)

    by Mawbid (3993) on Monday December 22, 2003 @11:17AM (#7786133)
    Much more interesting than SETI@HOME is the SIC@HOME [gamma.nic.fi] project, the search for incredible coincidences.

    A radio tuned to static is used to feed a stream of random data to a soundcard. The data is used to construct an image, and in the incredibly unlikely event that this image matches a predetermined image, you've proven that the universe is infinite! :-)

    Don't forget to check out the url of the "What is SIC@HOME?" page.


  • ad that we see alot of now -- is to BOINC her.

    Alas, nothing but fodder for the all the stalker fantasies of my fellow slashdotters.
  • by DeepDarkSky (111382) on Monday December 22, 2003 @11:34AM (#7786241)
    I thought a long time ago, why not make distributed computing applications as Java Applets hosted on web servers?

    Pros:
    - Nothing to "install".
    - Cross platform (write it once, run it everywhere, right?)
    - Easy to use (just browse)

    Cons:
    - Speed.
    - Full featured screen saver not possible?
    - uh...speed?

    • I agree with you and there already is a Java sub-culture doing just that - the Jini and JavaSpaces community [jini.org]. Highly distributed, self-healing, self-forming federations of services and distributed shared memory realms. Combine it with say Java WebStart [sun.com] for distribution and/or RIO [jini.org] for dynamic provisioning and you have one hell of a powerful distributed computing platform. And, because of the Java sandbox and the new Jini 2.0 security features, on that can be make sharing mobile code relatively safe. Throw in
    • There is nothing preventing a BOINC project from doing just that. If someone wants to make a Java (command line) port of SETI@home, feel free to send it to us.
  • Is that part and parcel of the overall design goal was to reduce cheating...

    Have we really fallen so far that people need to cheat on spare processor cycles donated?

  • This [berkeley.edu] is the BOINC Public License. IANAL, but at first read this looks very far from the GPL or LGPL... Anyone care to provide a better perspective on the legal issue?
  • Scientific progress goes BOINC?

    (according to Google this joke is not original, but what the hell)
  • For my Honors thesis, I produced a general-purpose platform-independent distributed computing system with the added benefit of presence awareness/work accounting. (As in it immediately reassigns your work unit when you go offline, rather than waiting indefinitely for you to return the results. This is reasonable because almost everyone who would run a distributed computing client has a 24/7 Internet connection.) See the PDF version of my thesis [theari.com] for more information.

Never test for an error condition you don't know how to handle. -- Steinbach

Working...