Data Analysis as a Service: An Infrastructure for Storing and Analyzing the Internet of Things
The Web of Things is an active research field which aims at promoting the easy access and handling of smart things’ digital representations through the adoption of Web standards and technologies. While huge research and development efforts have been spent on lower level networks and software technologies, it has been recognized that little experience exists instead in modeling and building applications for the Web of Things. Although several works have proposed Representational State Transfer (REST) inspired approaches for the Web of Things, a main limitation is that poor support is provided to web developers for speeding up the development of Web of Things applications while taking full advantage of REST benefits. In this paper, we propose a framework which supports developers in modeling smart things as web resources, exposing them through RESTful Application Programming Interfaces (APIs) and developing applications on top of them. The framework consists of a Web Resource information model, a middleware, and tools for developing and publishing smart things’ digital representations on the Web. We discuss the framework compliance with REST guidelines and its major implementation choices. Finally, we report on our test activities carried out within the SmartSantander European Project to evaluate the use and proficiency of our framework in a smart city scenario.