The research and development team of jobSkills.center has carried out long discussions on the database management system used by our group for our future projects.
Our activity required being able to store and process a lot of strongly connected data while maintaining very good performance.
Thus, the answer to this problem was to use a NoSQL database. This family of database management systems breaks away from the classic model of relational databases and emerged in the 2000s, meeting the needs of large companies that needed to handle colossal volumes of data.
The model of a graph database then appeared as the solution for our team. Among the graph database management systems, our team requested a partnership with Neo4j, which is positioned as an emerging product on the market.
It is an object-oriented database suitable for the exploitation of graph-type data structures, such as trees. In this model, data is represented by nodes and links. It works on the basis of the triplet: a subject representing the resource to be described, a predicate representing a type of property applicable to this resource and the object representing a data or another resource which is the value of the property. This model makes it possible to represent data in a natural way: jobSkills.center (subject) is a partner of (predicate) Neo4j (object).
Neo4j now allows jobSkills.center to model, store and process its data using all the power of its product. The lack of rigid modeling makes Neo4j well suited to handling changing data and frequently changing schemas. The associated query language is Cypher, specially developed for the purpose of making connected data processing operations simpler than SQL. The graph data model thus makes it possible to meet the expectations of new generation applications: performing complex queries while maintaining high performance.
Building on this partnership, jobSkills.center now exclusively uses this database management system. Indeed, thanks to the natural data modeling associated with the graph, the absence of a rigid diagram and the high performance of queries with the Cypher language, all our data is now stored in our unique Neo4j database, including rights management data and user account settings.