My friend Larry Braud, who is a Terry’s Computer Tips reader and a fellow volunteer instructor at the Cajun Clickers Computer Club, wrote me to suggest an article on one of the many distributed computing projects.
I have been participating in SETI@home, the earliest of the distributed PC computing projects, for over six years. This is the project that validated the usefulness of using hundreds of thousands of personal computers (or millions of them!) to run analyses of small chunks of data.
Almost all distributed PC computing projects are designed to run their analyses when the computer is otherwise idle, although some projects can be run all the time if the computer’s owner wants it that way. In effect, this widespread network on individual personal computers becomes a massively-powerful supercomputer.
The advantage of distributed PC computing is that the scientific project can (1) use the computers of volunteers without having to purchase or rent computer resources, and (2) with huge numbers of personal computers, they can actually have “systems” that are much more powerful than today’s fastest supercomputers.
I started several of my computers running the SETI@home project’s analysis package during April, 1999. Since then, I have had as many as four computers crunching data, and as few as one. SETI is the Search for ExtraTerrestrial Intelligence.
The SETI@home project analyses recordings of radio telescope data, looking for spikes and signal transitions that might indicate extraterrestrial life. When analysis of a data packet looks interesting, the scientists’ first step is to send the same data packet to someone else to make sure that the results are the same (yes, some people have attempted to fool the system). The Seti@home project is run from the University of California at Berkeley.
The original Seti@home project was run on custom-written software. After the resounding success of the project — in getting volunteers’ computers to run the project, not in finding ET — a more general distributed-computing software package was created called BOINC.
There are some major advantages to BOINC for distributed processing. First, in addition to the automatic submission of results and acquisition of new data to process (both of which were part of the original SETI@home software), BOINC allows the project to send automatic updates of both the BOINC software and any project-specific analysis procedures.
Did you notice that I’m referring to “the project” now and not SETI@home? That is because BOINC allows you, the individual computer owner, to decide which project you want to join, out of the many distributed computing projects being run.
The field of medicine is one of the major fields that is benefiting from distributed PC computing. Current BOINC-powered projects include Predictor@Home (biology), Rosetta@Home (biology), PrimeGrid (mathematics), SZTAKI Desktop Grid (mathematics), Einstein@Home (physics), LHC@Home (particle physics), SETI@Home (SETI), and Climateprediction.net (weather). BOINC is not the only game in town, either.
Another major medical project, Folding@home, is based out of Stanford University. This project is analyzing the process of protein folding, which is a critical and fundameental part of biology that is not yet understood. When proteins do not fold correctly (i.e. “misfold”), there can be serious effects, including many well known diseases, such as Alzheimer’s, Mad Cow (BSE), CJD, ALS, Huntington’s, Parkinson’s disease, and many cancers and cancer-related syndromes.
If you would like to help in any of these distributed computing projects, or just to read more about them, visit Folding@home or Seti@home. Both of these have good explanations of their processes, which are designed to be easily understood. Of course, either would be happy to let you download their free project software and start helping.