Querying in a Workload-Aware Triplestore Based on NoSQL Databases

RDF and SPARQL are increasingly used in a broad range of information management scenarios (e.g., governments, corporations, and startups). Scalable SPARQL querying has been the main issue for virtually all the recent RDF triplestores. This paper presents WA-RDF, a middleware that addresses workload-adaptive management of large RDF graphs. Our middleware not only employs all the most used NoSQL data models but also provides a novel RDF data partitioning approach based on a fragmentation strategy that maps RDF data into multiple NoSQL databases. This workload-aware partitioning scheme provides, in turn, efficient processing of SPARQL queries over these NoSQL databases. Our experimental evaluation shows that the solution is promising, outperforming three recent baselines.