CERN's Grid

I'm guessing this get some of you CSers out there salivating. The link below goes to a description of CERN's monster computing network called Grid. Unlike the Web, Grid not only shares information, but computing power and storage capacity. So instead of relying on ever bigger and faster computers to aggregate data and solve complex problems, they use the power of 2000 computers around the world connected on the CERN network. When the LHC is running at full force, it will be producing massive amounts of data and this is really the only way to handle it efficiently. Not only is it useful for storing data, but as you can see in the article, there are references to it being used to simulate drug/disease interactions to speed up drug development and the like. There are bunch of links off the site below that explain in more detail the processes behind the Grid like a real time map of the connections around the world. It's pretty badass.

http://public.web.cern.ch/public/en/Spotlight/SpotlightGrid-en.html

This type of grid computing has been around for quite some time. The major software package in this category is BOINC (Berkeley Open Infrastructure for Network Computing). This allows users all around the world to take on projects from universities and laboratories, and donate their computing power to help out these causes. The biggest project is probably SETI@home, which is the Search for Extra Terrestrial Life, but there are projects to cure diseases, study global warming, discover pulsars, and do many other types of scientific research.

There are currently close to 300,000 users worldwide, which is a pretty large network. It would be interesting to see a graph of the users as nodes and the connections are projects that each user is currently running. It would be a massive social network graph.

http://boinc.berkeley.edu/