A logic dimension on RDF partitioning, technical report

In the last years, scalable RDF processing systems distributing the data over a set of nodes to improve the performance have gained momentum. The triple is used as a distribution unit in these systems contrary to the relational model that defines the higher-level entities (tables) first and then partitions using tables’ subsets. We believe that gathering the triples storing facts of the same logical entities contributes not only to avoid scanning irrelevant triples but also to create RDF partitions with an actual logical meaning. In this study, we give the formal definition and detail the algorithm to gather the logical entities, which we name segments, used as distribution units for RDF datasets. The logical entities proposed, harmonize with the notion of partitions by instances (horizontal) and by attributes (vertical) in the relational model. We propose allocation strategies for these segments, considering the case when replication is available and in which both fragments by instances and by attributes are considered. We finally propose a declarative partitioning definition language for RDF declaring the higher-level entities and partitions.