Distributed Computing Applications of Context-Aware QoS and Trust Prediction Framework

Abstract

As more and more software systems are used in our day-to-day life from traveling, shopping to consuming health and financial services, our lives and habits are becoming highly dependent on the trustworthiness of these systems. Therefore, it is utmost important to develop techniques to: 1. Predict QoS and trust (i.e., the degree of compliance of a system to its specification) levels of systems at the early phases of system development lifecycle 2. Ensure such systems retain QoS and trust, while it is in production. However, it is a challenge to evaluate QoS and trust of systems with the changes in environment (which we refer as context) at the design time and the runtime of the system. We addressed this challenge by proposing a context-aware QoS and trust prediction model named ContextTrust model in our previous work. In this paper, we apply the ContexTrust model in following real-life applications: 1. Self-adaptations of system with degradation of QoS and trust at runtime, 2. Decomposition of QoS and trust of systems, 3. Prediction of QoS and trust at design-time for selected context. The results indicate the prediction model is useful in variety of applications that help to build QoS-optimized trustworthy distributed systems.