|
Runtime Model Checking for SLA
Compliance Monitoring and QoS Prediction University of Naples Parthenope, 80143
Napoli, Italy Abstract Sophisticated workflows, where multiple parties cooperate towards the achievement of a shared goal are today common. In a market-oriented setup, it is key that effective mechanisms be available for providing accountability within the business process. The challenge is to be able to continuously monitor the progress of the business process, ideally, anticipating contract breaches and triggering corrective actions. In this paper we propose a novel QoS prediction approach which combines run-time monitoring of the real system with probabilistic model-checking on a parametric system model. To cope with the huge amount of data generated by the monitored system, while ensuring that parameters are extracted in a timing fashion, we relied on big data analytics solutions. To validate the proposed approach, a prototype of the QoS prediction framework has been developed, and an experimental campaign has been conducted with respect to a case study in the field of Smart Grids. Keywords: Big Data
Analytics, QoS Prediction, Model Checking, SLA
compliance monitoring +: Corresponding author: Luigi Coppolino Journal
of Wireless Mobile Networks, Ubiquitous Computing, and Dependable
Applications (JoWUA), |