Quantitative Verification for Monitoring Event-Streaming Systems

High-performance data streaming technologies are increasingly adopted in IT companies to support the integration of heterogeneous and possibly distributed applications. Compared with traditional message queuing middleware, a platform enables implementation event-streaming systems (ESS) which include not only complex queues but also pipelines that transform react streams data. By analysing centralised streams, one can evaluate Quality-of-Service for other components produce or consume those streams. We consider exploitation probabilistic model checking as performance monitoring technique ESS systems. Probabilistic checking is mature, powerful verification successful application analysis. However, an system may contain quantitative parameters determined by event observed certain period time. In this paper, we present novel theoretical framework called QV4M (meaning “quantitative monitoring”) systems, based on two recent methods probabilistic checking. assumes random variables infers statistical significance output. empirical evaluation computational time cost QV4M.