parasitic computing

(pair.uh.SIT.ik kawm.pyoo.ting)

n.

Using a series of remote computers to surreptitiously perform calculations that are part of a larger computational problem.

Example Citation:

"Four researchers at the University of Notre Dame have figured out how to use ordinary communications among machines on the Internet to trick Web servers into solving math problems for them. ... Although parasitic computing does not compromise the security of the host, according to the researchers, since it only uses normal Internet communications, it can impact the server's performance. This could turn into a sort of denial of service attack."

— Lynn Greiner, "Infested computers raise ethical questions," Computing Canada, September 21, 2001

Earliest Citation:

"Reliable communication on the Internet is guaranteed by a standard set of protocols, used by all computers. Here we show that these protocols can be exploited to compute with the communication infrastructure, transforming the Internet into a distributed computer in which servers unwittingly perform computation on behalf of a remote node. In this model, which we call 'parasitic computing', one machine forces target computers to solve a piece of a complex computational problem merely by engaging them in standard communication."

— Albert-Laszlo Barabasi, Vincent W. Freeh, Hawoong Jeong, and Jay B. Brockman, "Parasitic computing," Nature, August 30, 2001

Notes:

Unlike computational grid setups (such as the SETI\@home project: http://setiathome.ssl.berkeley.edu/) where people voluntarily offer their computers' idle time to crunch numbers, parasitic computing operates by stealth. It takes advantage of a standard computation — called the checksum by communications nerds — that every Web server on the Internet normally uses as a matter of course to verify that incoming data hasn't been corrupted. Apparently it's possible to take these checksum calculations and use them to solve a tiny piece of a larger problem. Do this with enough computers (the Internet has millions of servers) and you can solve the larger problem in its entirety.

That's the good news. The bad news is that, although messing with a few checksums won't cause a perceptible drop in the performance of the server, hijacking millions or billions of checksum calculations would bring the machine to its digital knees. It's just one more thing to keep Web site administrators chewing their fingernails.

Related Words:

computational grid

distributed denial of service

man in the middle attack

Categories:

Hacking and Hackers

Internet