An architecture to analyze big data in the Internet of Things

Abstract

Internet of Things (IoT) is nowadays increasingly becoming a worldwide network of interconnected devices uniquely addressable, via a standard communication protocol. Such devices generate a massive volume of heterogeneous data, which lead a system towards a major computational challenges, such as aggregation, storing, and processing. Also, a major problem arises when there is a need to extract useful information from this massive volume of data. Therefore, to address these needs, this paper proposes an architecture to analyze big data in the IoT. The basic concept involves the partitioning of dynamic data, i.e., big data with the complex magnitude is divided into subsets. These subsets are based on the theoretical model of data fusion, which works in the Hadoop processing server to enhance the computational efficiency. The proposed architecture is tested by analyzing healthcare data sets, mainly comprises of activities including walking, running, ECG. The feasibility and efficiency of the proposed architecture are implemented on Hadoop single node setup on UBUNTU 14.04 LTS core™i5 machine with 3.2 GHz processor and 4 GB memory. The results show that the proposed architecture efficiently analyze the massive volume of data with a maximum throughput.