The CPS with the Hadoop ecosystems

The cyber physical systems (CPS) are timing sensitive with the real-time constraints. Under the industrial domain, the CPS generates a huge amount of real-time data, which gives essential patterns of system behavior and safety-critical logs. In this paper, we introduce system architecture for the CPS with the Hadoop ecosystem. With the characteristics of CPS processing, the huge amount of data and its processing is not able to be analyzed on a single machine. Instead, we propose system architecture to store and process the RT-based generating data with the Hadoop ecosystem. The Hadoop Distributed File System (HDFS) is used as a fundamental DFS system for storing the huge CPS data, the Hive is used to analyze the CPS data for data warehousing processing, the Flume is used for collecting data from servers and processing with the HDFS, and the RHive is used for applying data-mining and analysis. We introduce the architecture overview and system design strategies to work on an efficient performance and achieve the effective data analysis.