Knowledge-based distributed conflict resolution for multiparty interactions and priorities

Abstract

Distributed decentralized implementation of systems of communicating processes raises non-trivial problems. Correct execution of multiparty interactions, subject to priority rules, requires sophisticated mechanisms for runtime conflict detection and resolution. We propose a method for detection of false conflicts which combines partial observation of the system's state and apriori knowledge extracted from invariants. We propose heuristics for determining optimal sets of observations leading to implementations with particular guarantees. We provide preliminary experimental results on an implementation of the method in the BIP framework. © 2012 IFIP International Federation for Information Processing.