PROJECT TITLE:

Diplo cloud Efficient and Scalable Management of RDF Data in the Cloud - 2016

ABSTRACT:

Despite recent advances in distributed RDF Knowledge Management, processing large-amounts of RDF information in the cloud remains very difficult. In spite of its seemingly simple knowledge model, RDF actually encodes wealthy and complex graphs mixing each instance and schema-level data. Sharding such information using classical techniques or partitioning the graph using traditional min-cut algorithms ends up in very inefficient distributed operations and to a high range of joins. In this paper, we have a tendency to describe DiploCloud, an efficient and scalable distributed RDF knowledge management system for the cloud. Contrary to previous approaches, DiploCloud runs a physiological analysis of both instance and schema data prior to partitioning the information. In this paper, we have a tendency to describe the architecture of DiploCloud, its main information structures, additionally because the new algorithms we use to partition and distribute information. We have a tendency to also gift an extensive analysis of DiploCloud showing that our system is typically 2 orders of magnitude faster than state-of-the-art systems on normal workloads.