Implementation of docker-based smart greenhouse data analysis platform

Through the convergence of ICT-technology with traditional greenhouse, the smart greenhouse has been able to reduce costs and increase production by using the large amount of data for making the optimal growth environment. Korean agricultural agency EPIS provides the raw data that is gathered from the greenhouses to public users using open API. In order to get an optimal growth algorithm, it needs to build up a platform for data analysis with various combination of factors easily. This paper describes the experience of implementing a platform that can easily analyze farm environment information. This platform consists of data gathering engine, data visualizer, database, and traffic router. We have implemented data gathering engine with Python to fetch the data using SOAP-based Open API, and uses NoSQL time-series database for storing them, open sources server applications for visualization and traffic routing. Furthermore, this application servers are all running as a container. The use of Docker makes it much easier to deploy this platform services regardless of underlying operating systems. In this paper, we describes the implementation of data gathering engine and how we composed the micro services to build up agricultural data analysis platform.