A dependency matrix based framework for QoS diagnosis in SOA

When a business process fails to deliver the quality of service (QoS), it is necessary to identify the faulty services that cause the problem. In this paper, we present a framework for diagnosing QoS problems in SOA. A set-covering algorithm is used to select the locations of run-time service data collection, or probes. The framework then creates a dependency matrix to denote the relationships between data recorded by probes and service status. A polynomial time diagnosis algorithm based on the dependency matrix and the probe data is then used to identify potential faulty services. The algorithm also uses predicates on probes (PoP) to compare the data reported from neighboring probes. The diagnosis framework does not require historical knowledge on services. Simulation results show that the diagnosis framework is both effective and efficient.