부산대학교 도서관

Data platforms with an ability to capture, process and analyze high frequency streaming data from vast complex systems reliably with high scalability is the central problem in this paper. Particularly when developed data platforms are to be deployed within extreme climatic conditions. We have developed a decoupled data pipeline by fusing multiple services within edge and cloud computing paradigms following design principles in software reliability to provide optimal services during vast human studies in extreme climatic simulations. The data pipeline is demonstrated through a pre-deployment acclimation case study in climatic chambers. Performance evaluation of each service within the developed pipeline shows high levels of reliability and availability across multiple testing and study simulations. As a result, real-time data from multiple complex data sources are made available for on time personalized analytics for analysis of physiological responses to austere environmental conditions.