Distributed solutions to the mutual exclusion problem

The authors propose the token path method for achieving mutual exclusion in distributed computing systems. The nodes in the distributed system interact only with their neighbors. There is no central control exercised. The node which wishes to execute the critical section perturbs the existing token path and waits for the token. When the token arrives at the node, the processes which request control for executing the section are given permission to execute the section according to some local policy. After that, the token is routed to another node which has requested the token. All messages are routed on the edges of the spanning tree of the system, which helps to reduce the number of messages. The nondeterministic selection of one-of-many critical sections is also presented, and a solution using the token path method is proposed