Quote:
| Originally Posted by tejas Ok..I will try to explain in simpler language!!
When you are using computer, not 100% of your CPU is utilized. Any OS you use, all the processes and applications don't use all your CPU power and System Idle Process uses a huge share without giving you any result.
Distributed Computing is a project where instead of having centralized computer labs, people are encouraged to participate in the project and the results of simulation is processed at one place.
Folding at Home, is one such project.(Imagine, how much it would cost to have millions of systems at one place to run simulations.) So, the client you install on your system runs the simulations and finally when one Work Unit is finished, the results are sent back to Stanford labs where they are processed, analyzed.
The simulation are run to utilized the share which System Idle Restore process eats away. All the work you do carries some points, so more your computer contributes, more points you get.
It is easy, installation free! Even if you don't have admin rights on a computer, you can run it! Internet connection is not absolutely necessary. You can run simulations without it and then it is only needed for sending results and receiving new Work Units. |
Among the many distributed computing projects out there are seti@home and folding@home.
Seti@home is probably the most well-advertised and thus well-known programs out there, where one downloads a screensaver program that analyses radio signals captured from space for artifically created patterns which would indicate another intelligence in the universe.
Folding@home is a similar project, where one can download a screensaver that performs protein-folding simulations-- proteins are tremendously complex groups of amino acids which can fold themselves into many shapes, and it takes a lot of computing power to simulate that folding action for any particular protein. By distributing a folding sequence among thousands of computers worldwide, the scientists who need the data to figure out what the proteins do and whether this experimental drug or that is likely to do its intended job, can obtain results in a lot less time than if they had to wait for a chance on a supercomputer or two and wait for that it to do the sequence.
Stanford also provides a console-only client which concentrates on doing just the folding calculations without the overhead required to make the folding visualizations for the screen saver (the screen-saver version performs the visualizations regardless of whether they are visible or not). Those of us who are "serious" folders prefer to use the console-only client because that allows our computers to perform more folds than the screensaver would for a particular period of time.
In all cases of distributed computing, the client only runs when the computer is idle. The F@H client can be set to use as little or as much of the CPU's processing power as you wish, can be set to automatically dial out to send its results and download new tasks or to wait for you to give it permission to access the web. It can shut itself off if the laptop is on battery power, it can load as a service in WinXP so the process is absolutely transparent (and no one has to be logged in).
F@H simulates protein folds, this data is used by biochemists and medical scientists to investigate how drugs perform their jobs, how cancers, viruses and pathogens infect and spread within a host and investigate ways of preventing it.
The Stanford site has a lot of information on exactly what the current projects are for and what they have accomplished so far with the program.
Dev Articles Community Forums Sponsor:
October 25th, 2005, 12:00 PM
right
Del.icio.us
Digg
Google
Spurl
Blink
Furl
Simpy
Y! MyWeb
