5.8. Conclusion
In this chapter, a multigraph-based engine AMBER has been proposed in order to answer SPARQL queries. The multigraph representation has given us the following two advantages: (1) it enables us to construct lightweight indexing structures that ameliorate the time performance of AMBER; (2) the graph representation itself motivates us to exploit recent graph management techniques. The proposed engine AMBER has been tested over large RDF triplestores. We have observed that AMBER outperforms its state-of-the-art competitors on two main aspects: (1) its robustness with respect to the query size and (2) its excellent time performance.
