Searching for an Optimal Data Platform for Relevant Information Search in Inorganic Chemistry and Materials Science

Choosing the most suitable database management system is one of the most critical challenges in developing any information system operating on big data. When selecting, as a rule, the overall system speed is considered the main criterion regarding certain data structures due to the subject area specifics. In the current article, using the example of searching for relevant information on inorganic compounds, an attempt is made to analyze the possibility of using relational and graph database management systems (DBMSs) to build a data storage subsystem. Graph-based database implementations, powered by SQL Graph and Neo4j, are considered and compared with a relational version based on SQL Server. Typical query execution speed comparative analysis is carried out when searching for relevant information in the field of inorganic chemistry and materials science. It’s shown that, due to the graph nature of relevance definition, graph DBMS outperforms relational.