Applying a Path-Compression technique to Obtain an Efficient Distributed Mutual Exclusion Algorithm

In this paper we present a distributed algorithm for mutual exclusion. The algorithm maintains a dynamic forest structure in which the paths between nodes are compressed as a result of requesting the Critical Section. We develop a formal model of the algorithm's execution, which enables us to prove its correctnes. The formal model is also used to show that an execution history of the algorithm when concurrent requests are made (the ususal case) is equivalent to a history in which the requests are made serially. Based on this fact we are able to prove a logarithmic upper bound on the average number of messages needed per critical section grant.