Using Markov-Chains to Model Reliability and QoS for Deployed Service-Based Systems

An approach for modeling Service-based Systems (SBS) is proposed. The approach is intended as an aid for estimating the reliability and Quality of Service (QoS) attributes of an SBS against changes in its atomic services over a deployment period. A Markov-chain model is used to formally capture the status and evolution of an SBS. Service selection policy is considered in the model of the user operational profile to account for its impact on the reliability and QoS of an SBS. Both static and dynamic analysis techniques are used to obtain the estimates of reliability, availability, and the average response time of an SBS. Simulation studies aimed at investigating the effectiveness of the proposed approach are reported.