Data models in NoSQL databases for big data contexts

Abstract

Data models are a central piece in information systems, being the relational data models very popular and extensively used. In Big Data, and due to the characteristics of the NoSQL databases, the data modeling task is seen in another perspective, as those databases are considered schema-free. Nevertheless, these databases also need data models that ensure the proper storage and querying of the data. Considering the vast amount of relational databases and the ever-increasing volume of data, the importance of data models in Big Data increases. In this work, a specific set of rules is proposed for the automatic transition between a traditional and a Big Data environment, considering two specific objectives: the identification of a columnar data model for HBase supporting operational needs and the identification of a tabular data model for Hive supporting analytical needs. The obtained results show the applicability of the proposed rules and their relevance for data modeling in Big Data environments.