Parasitic computing

From Wikipedia, the free encyclopedia

Parasitic computing is a computing technology where a remote computer tricks a target computer into performing computations of a complex nature under disguise of a standard communications session. For example, this can occur as follows: when a request is sent to a web server for a web page, the communication programs in the requesting machine split it into information packets before sending it across the Internet. When these packets reach the target machine, they go through certain communication layers before reaching the target program--here, the web server that serves the requested page. One such layer is the TCP (transmission control protocol) that assembles the packets in proper order and makes sure that all the packets are in proper shape before handing them over to the web server. During this stage the TCP component makes some computation to ascertain the validity of the received information packets. As reported in the magazine Nature, this aspect of the TCP, called a checksum, is exploited to deploy the computing power of various servers (without any permission) by a few scientists to make some computations and thereby convert the Internet into a giant distributed computer in which the servers perform computation on behalf of a remote node.