Self-stabilizing deadlock algorithms in distributed systems

A self-stabilizing system is a network of processors, which, when started from an arbitrary (and possibly illegal) initial state, always returns to a legal state in a finite number of steps. Self-stabilization is an evolving paradigm in fault-tolerant computing. This research will be the first time self-stabilization is used in the areas of deadlock detection and prevention. Traditional deadlock detection algorithms have a process initiate a probe. If that probe travels around the system and is received by the initiator, there is a cycle in the system, and deadlock is detected. In order to prevent deadlocks, algorithms usually rank nodes in order to determine if an added edge will create a deadlock in the system. In a self-stabilizing system, perturbances are automatically dealt with. For the deadlock model, the perturbances in the system are requests and releases of resources. So, the self-stabilizing deadlock detection algorithm will automatically detect a deadlock when a request causes a cycle in the wait-for graph. The self-stabilizing prevention algorithm prevents deadlocks in a similar manner. The self-stabilizing algorithms do not have to be initiated by any process because the requests and releases create a perturbance which is dealt with automatically