A categorization of RDF triplestores
Khadija Alaoui · Proceedings of the 4th International Conference on Smart City Applications · 2019
The management of data produced by multiple services of our daily life necessitates to be done intelligently to come up with smart services and intelligent answers to the challenges our world is facing. RDF has proven to provide a good solution for the data modeling in various application fields. Various RDF data management systems with various characteristics have been proposed to handle RDF. For each specific enterprise application domain, it is therefore a necessity to choose the adequate system that helps to deal with associated data in an efficient way. To help enterprise applications developers in such task, we give a categorization of the existing RDF management systems, also called triplestores, according to their main characteristics. The categorization is motivated by the reasons behind the adoption of one or the other system for handling RDF data. Existing categorization works only treat a limited number of categories related to a relational or a native database classification and consider only a small number of triple stores. However, our categorization takes into account various factors related among others to RDF objectives, data organization, query processing, dynamicity and scalability and in accordance with the diverse developments in data modeling and manipulation technologies. This allows users to choose the most appropriate RDF system depending on their needs.