In-Memory Data Grid System for Real-Time Processing of Machine Sensor Data in a Smart Factory Environment
Han-Sol Park, Jin‐Hyuk Kim, Chi‐Hwan Choi, Bo-Ra Jung, Kyung-Hee Lee, Suyoung Chi, Wan-Sup Cho · 2015
Industry 4.0 is aimed at setting up a smart factory, which focuses on developing base technologies such as Internet Of Things (IOT), sensor, cyber-physical system and etc. The smart factory produces process data in real time through the sensor for each process. Thus, it is required to ensure simultaneous processing of large amounts of data poured in real time from sensors. Since a traditional RDBMS suffers from the degradation of operation performance and requires more CPUs and memories as the amount of data to be processed increases, the costs for configuration in parallel by adding servers is also increased[3]. However, if in-memory technology is applied, it is possible to process large amounts of data in real time. In this regard, this study seeks to design and implement a system that can apply Hazelcast, which is one of in-memory data grids (IMDG) that extends in-memory technology, to the smart factory. Lastly, it proceeds with concurrency test in a situation of multi-connection between the existing RDBMS and IMDG.