How Does This All Work??
17 Aug 2005 19:26:00 UTC
Topic 189709
(moderation:
I am just curious of how this was all started. I read a little bit in "Scientific American" but in awe of how this all works. I know that it uses the 'idle time' from each computer; I can honestly say that I have no idea what that really is.
In short, I am a amateur. Can someone please bring me up to speed with everything. I feel as if I have been living under a rock.
How Does This All Work??
)
Something like this...
You tell your computer to do something, it does it, and then it waits for you to tell it to do something else...
... except, it does not sit and wait, it is running at full speed waiting for something to do...
... so your next command comes and it processes it...
... in the mean time it has been "idle" doing nothing for seconds, with todays machines, those seconds represent a vast amount of wasted power - they could have been doing something rather then waiting...
... so along come wasted cycle harvesters like BOINC that allow your machine to actively be doing something when you are not.
It gets a lot more detailed, but that is the essence.
Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.
Wow. OK, I'll give it a
)
Wow. OK, I'll give it a try...
When you aren't using your computer, you just walk away and leave it turned on, it is still "doing something". That could be running a pretty screensaver, or just sitting twiddling it's electronic thumbs going "has he hit a key? moved the mouse? no? OK. has he hit a key? moved the mouse? no? OK..."
Around 1999, the folks at U.C.Berkeley said "Why don't we give a few hundred of those PCs some software that will not only be a pretty screensaver, but will also DO something? We have all this telescope data coming in faster than our computers can crunch it..." Instead of a few hundred, it hit the news, and every science fiction fan and computer nerd in the semi-civilized world suddenly joined SETI, looking for ET. A single "work unit" from the telescope could be crunched on the average PC in a couple of days of "spare time", and the results automatically sent to UCB and the next work unit downloaded. As PCs got faster, UCB added more work to each work unit, but even so, today the "average" PC can crunch several SETI work units in a day.
Because some computers are faster than others, the simple "count of how many work units someone has done" became a competition, with teams, certificates, bragging rights...
Other projects thought "We can spend $10,000,000 on a supercomputer... or get together with UCB and let all those volunteers do the work for us!", and thus was born BOINC, to handle mutliple projects (and divide the credits awarded fairly), not just SETI. (The volunteers said "no way" to doing anything that was "for profit", so the various projects are all non-profit science-related in some way or another.)
Newer "energy efficient" computers go into a "sleep" mode to save electricity, effectively shutting themselves down so there is no real "idle time" anymore. So volunteering today isn't "free", it WILL cost you a bit in electricity and wear and tear on your computer, but generally both are very minimal. You are making a contribution to scientific research, and can select which projects you are interested in contributing to. Plus, at most projects, you can get a pretty certificate saying how many credits you've earned! (Obligitory Star Trek reference: Credits are directly exchangeable for slips of gold-pressed latinum at any Ferengi bank. Watch out for exchange fees.)
Hope that helps! :-)
Here is a link to the
)
Here is a link to the Unofficial Boinc WIKI. It has a wealth of information. It is a work in progress done entirely by volunteers. If you have any questions / comments / or suggestions please make them known. Your questions will be addressed.
Jim