Optimization of RDF Data Preprocessing for METIS Partitioning

The Resource Description Framework (RDF) developed by W3C is increasingly adopted to model data in a variety of scenarios, especially the published or exchanged data on Web. Managing a large volume of data is difficult because of size, heterogeneity, and additional complexity brought by the RDF reasoning. To take up the challenge of the massive size of RDF data, storage architectures and distributed processing are required. For this, the partitioning of data set is a widely adopted technique in many systems to deploy easily the distributed and parallel architectures. We study in this paper, the reduction of the dataset size to be partitioned into a step called preprocessing by improving data partitioning.