A Petri net representation of the Take-Grant model

The Take-Grant model is formalised with a Petri net notation. It is shown that this approach offers a convenient way to deal with the problem of determining the cooperation required to share or steal rights. An algorithm is proposed that finds all rights that a subject can steal with the help of a given set of conspirators. With the Petri net abstraction, the problem can be solved in linear time. It is compared to another algorithm exposed in the literature that requires n/sup 2/ operations for a n subject graph. >