Self-stabilizing distributed algorithms for acyclic graphs
Abstract
A self-stabilizing distributed system is a network of processors, which when started from an arbitrary and possibly illegal state, always returns to a legal state in a finite number of steps. Two self-stabilizing protocols for distributed systems are presented in this thesis. The first protocol topologically sorts the processors in a distributed system of directed acyclic graph (DAG) topology and uses this information to build a shortest path routing table in each node in the system to all accessible nodes from that node. The second protocol determines the rank of the individual processors in a distributed tree network based on the values possessed by them. Due to the self-stabilizing nature of these protocols the system can withstand transient errors and recover automatically from them.