Design And Implementation Of Real-time Data Processing Platform Based On Event Driven Architecture

The rapid development of IoT(Internet of Things)not only brings great convenience to people,but also produces a large amount of sensor data with potential value.Sensor data is massive,diverse,heterogeneous and dynamic.How to make better use of massive sensor data is one of the hotspots in the field of Internet of Things.EDA(Event Driven Architecture)is loosely coupled,highly integrated,initiative and real-time,which meets the requirements of IoT applications.Currently,the IoT services platform requires an efficient and highly usable real-time data processing subsystem,so the paper focuses on combining EDA and Spark engine to build a loosely decoupled,expansible,high-performance data processing platform and designing user-friendly graphical interfaces to simplify the operations of data process modeling.The work details are as follows:Firstly,a data access module is designed to obtain events from the publish/subscribe system and pretreat them.Secondly,in order to deal with the problem of complex data process modeling,this paper designs a graphical module of data process modeling.By using it,users can model the data process simply.Thirdly,this paper implements user-friendly interfaces to help people configure the platform and view the results of data processing easily.In addition,the platform integrates Spark engine to improve the calculating speed.Finally,we do corresponding tests and the results show that this platform can meet our expectations.The real-time data processing platform has the characteristics of easy management,simple data process modeling,good real-time processing performance,event driven architecture and so on,which meets the needs of the IoT services platform.