Folding-community.org ( http://forum.folding-community.org/homepage.php ) forums would be a good place to start figuring this information out. I did some quick searches, and the moderators appear to be very willing to help people understand how F@H works.

Basically, F@H gets data from stanford servers like a web browser, using ports 80 and 8080. I would assume that HTTP is used. It only connects when it completes a work packet/protien/etc and uploads the results of the simulation to the servers and gets a new work packet. The F@H client does not open any ports on your computer to listen, it only connects to stanford's servers for workpacket stuff.

I don't know of any vulnerabilities in F@H. From what I understand of it, it uploads simulation results to stanford. It then downloads a new simulation. The stanford servers pick the best simulation for the speed of the computer (the F@H client reports benchmark/CPU speeds) so the work packet finishes simulations in time. The work packet is basically a text file, with a LOT of numbers that have to do with positions and temperatures of different atoms that make up protiens. It then probably verifies the files, and if something is wrong, it would probably stop. You can ask the above posted community about it, along with searching for posts related to "Security" on their forums.



Interesting Note: AntiOnline is in position 404. It correctly describes what happened to the webpage we had, and participation.

Another Note: I don't fold much anymore. My power-hungry computer is now being turned off...